Focused on accounts and posts that discuss concrete model performance: benchmarks, evals, leaderboards, coding-agent scores, open-weight comparisons, cost/speed/quality, multimodal and scientific-model performance. Broad celebrity/futurist accounts are excluded.
No Elon-style mega-celebrity list here: this table is for people/accounts posting concrete model-performance evidence.
Independent model quality, intelligence, speed, price, cost-per-task, and coding-agent comparisons.
Why included: Independent model quality, intelligence, speed, price, cost-per-task, and coding-agent comparisons.
Publishes Chatbot Arena / model leaderboard results and arena methodology.
Why included: Publishes Chatbot Arena / model leaderboard results and arena methodology.
Originator of Chatbot Arena/FastChat-style model evaluation work.
Why included: Originator of Chatbot Arena/FastChat-style model evaluation work.
Maintains live/refreshing benchmark results to reduce contamination and stale evals.
Why included: Maintains live/refreshing benchmark results to reduce contamination and stale evals.
Publishes practical coding model benchmark results from Aider workflows.
Why included: Publishes practical coding model benchmark results from Aider workflows.
Creator of Aider; frequent publisher/interpreter of coding model leaderboard results.
Why included: Creator of Aider; frequent publisher/interpreter of coding model leaderboard results.
Official software-engineering benchmark and leaderboard for AI coding agents.
Why included: Official software-engineering benchmark and leaderboard for AI coding agents.
Developer-facing model release and performance details.
Why included: Developer-facing model release and performance details.
Publishes model rankings across legal, financial, coding, multimodal, and other private benchmarks.
Why included: Publishes model rankings across legal, financial, coding, multimodal, and other private benchmarks.
One of the highest signal technical AI explainers on X.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Major applied-AI educator and practitioner influence.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-authority AI-for-science and AGI lab voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Major technical counterweight to LLM-scaling consensus.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Rare posts carry frontier-lab/research weight.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
AI search and consumer-AI distribution leader.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Major AI research and human-centered AI authority.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Important benchmark/reasoning voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-authority voice on deep learning and AI risk.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Claude/frontier AI strategy and safety framing.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Elite systems + ML research credibility.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Open-source AI distribution and community leader.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Data, evals, and enterprise AI influence.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-signal physical AGI and robotics commentator.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Hands-on LLM training and architecture education.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
PyTorch ecosystem credibility.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Widely cited technical explainer.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Applied technical AI education and open models.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Open-source AI and generative media commentary.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Developer-facing AI launch signal.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Top practical technical evaluator of LLM tools.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-signal practitioner voice on AI evals.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Useful model-design and frontier-lab research signal.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Foundational ML researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Major DeepMind technical leader.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-credibility research voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Applied ML competition and tooling expert.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Influential skeptical AI commentator.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Practical AI/data app educator.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Longstanding ML research figure.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Recognized ML academic voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Large practical ML learning audience.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Applied ML/Kaggle KOL.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Large applied AI education audience.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-quality ML research voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical computer-vision KOL.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Practical ML engineering audience.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Authoritative ML textbook/research voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical ML theory voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Important efficient/open AI researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Foundational ML researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Strong ML educational voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Rising technical AI research voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical DeepMind/social-good voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical ML/statistics KOL.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical ML academic.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Sharp technical ML critic.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Senior ML research authority.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical RL/agents researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Important technical retrieval/embedding voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Hands-on ML builder.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical ML/statistics voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Turing laureate and major AI safety/research voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Core interpretability KOL.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical robotics/HRI authority.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Major robotics/RL KOL.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-signal robotics/learning researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Top technical robotics/RL researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Important technical interpretability voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Senior deep learning researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical NLP researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical ML/science-AI voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical and creative AI research voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Foundational sequence-model researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Senior technical Google AI researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Senior AI/robotics authority.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical founder with model lineage.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical educator and fast.ai cofounder.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Core open-source AI platform builder.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Important open-source AI infra voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Robotics/AI education pioneer.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Strong responsible-AI researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical/public-interest AI voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical and skeptical AI researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Popular practical ML educator.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-signal AI engineering voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical data/AI product KOL.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Important AI auditing voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical AI and policy critic.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-signal practical technical educator.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical responsible-AI voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Top technical alignment researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Core technical alignment voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Top LLM-evaluation researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical NLP/agents researcher.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Important eval/safety benchmark creator.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical theory/safety voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Top AI engineering community builder.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-signal AI startup/model curator.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical AI product/research voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Strong AI interview/eval commentator.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Top technical post-training/open-model voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Raw research discovery channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
AI paper discovery/amplification.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-signal technical AI explainer.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Practical model builder.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Primary OpenAI launch channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Primary Anthropic launch channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Major technical AI lab channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Key open-model lab channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
European frontier model channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-impact open AI distribution.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Enterprise LLM platform channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Open AI infrastructure channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
xAI primary launch channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Useful model market/API signal.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Model deployment ecosystem channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
MLOps and AI eval community.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Key AI agent/RAG developer channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Important AI engineering framework.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Technical AI infra channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Important scalable AI systems channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Evals/data infrastructure influence.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-signal eval/benchmark channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Important agent-eval methodology voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Government eval-framework voice.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Bio/chem/cyber eval relevance.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Methodology/benchmark analysis amplifier.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Standards-setting benchmark channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Broad benchmark visibility channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Foundational drug-discovery benchmark ecosystem.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Direct DDDBench peer/partner target.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-signal biomedical-agent benchmark KOL.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Top scientific-agent benchmark target.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Relevant industry-standard DD benchmarking channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
High-credibility DD benchmark evaluator.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Community amplifier for open BioML benchmarks.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Strong AI-bio benchmark amplifier.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Relevant chemistry benchmark group.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Strong chemistry/science-agent benchmark KOL.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Chemistry model/eval KOL.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Protein/structure benchmark channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Protein-design benchmark credibility network.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Benchmark/data community account.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Medical LLM benchmark KOL.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Generative AI model distribution channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Major creative AI model channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Major generative video model channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Creative model product channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Major AI coding tool channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Important coding-agent benchmark/product channel.
Why included: Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence.
Separate table for recent model-performance posts. High/Medium rows are direct post URLs; Low rows are X search links for follow-up discovery/API verification.
GPT-5.6 benchmark launch analysis
Models: GPT-5.6 Sol/Terra/Luna; Claude Fable 5
Benchmarks: Coding Agent Index; DeepSWE; Terminal-Bench v2; SWE-Atlas-QnA
Sol leads coding-agent evals and approaches Fable intelligence at lower cost
Engagement: n/a · direct X result
GPT-5.6 public release benchmark claims
Models: GPT-5.6 family
Benchmarks: Agents Last Exam; Terminal-Bench; HealthBench; AA Index
Primary model-release post for latest OpenAI benchmark family
Engagement: n/a · subagent/web search
Developer/API performance for GPT-5.6
Models: GPT-5.6 family
Benchmarks: coding; tool use; context/pricing
Builder-facing model-performance details
Engagement: n/a · subagent/web search
Grok 4.5 independent benchmark
Models: Grok 4.5; GPT-5.5; Fable 5; Opus 4.8
Benchmarks: AA Intelligence Index; Coding Agent Index
Places Grok 4.5 high overall and efficient for coding agents
Engagement: n/a · subagent/web search
GPT-5.6 Sol Terminal-Bench SOTA
Models: GPT-5.6 Sol
Benchmarks: Terminal-Bench 2.1
OpenAI says Sol sets SOTA on complex command-line workflows
Engagement: n/a · web_search exact X result
Capability indices by domain
Models: Fable 5; Opus 4.8; GPT-5.5; Gemini 3.5 Flash; GLM-5.2
Benchmarks: AA Capability Indices
Domain-specific model ranking shifts and open-weight leaders
Engagement: n/a · subagent/web search
Cost-per-task frontier comparison
Models: DeepSeek V4 Flash; GLM-5.2; frontier models
Benchmarks: AA Strategy & Ops Index; cost/task
Highlights cheap open-weight task completion vs frontier premium
Engagement: n/a · subagent/web search
DrugDiscoveryBench launch
Models: GPT-5.5; Gemini 3.5 Flash; Opus 4.8
Benchmarks: DrugDiscoveryBench
Independent benchmark for early drug-discovery computational work
Engagement: n/a · subagent/web search
DrugDiscoveryBench summary
Models: leading AI agents
Benchmarks: DrugDiscoveryBench
Broad amplification of life-science model benchmark
Engagement: n/a · subagent/web search
LifeSciBench release
Models: GPT-Rosalind; GPT-5.5
Benchmarks: LifeSciBench
Life-science benchmark: GPT-Rosalind improves reported performance
Engagement: n/a · subagent/web search
Claude Fable 5 tops AA Intelligence Index
Models: Claude Fable 5; Opus 4.8
Benchmarks: AA Intelligence Index
Fable 5 score/rank post from independent benchmark source
Engagement: n/a · subagent
Fable 5 knowledge/hallucination eval
Models: Claude Fable 5; Opus 4.8
Benchmarks: AA-Omniscience
Knowledge reliability / hallucination-rate comparison
Engagement: n/a · subagent
Fable 5 benchmark cost
Models: Claude Fable 5; Opus 4.8; GPT-5.5
Benchmarks: AA Intelligence Index; cost/task
Shows benchmark-run cost and premium frontier pricing
Engagement: n/a · subagent
DeepSWE replaces SWE-Bench Pro
Models: GPT-5.5 Codex; Claude Code Fable 5; Opus 4.8
Benchmarks: DeepSWE; SWE-Bench Pro
Coding-agent benchmark contamination/gameability discussion
Engagement: n/a · subagent
SWE-Bench Pro anomaly
Models: GPT-5.5 Codex; Opus 4.8
Benchmarks: SWE-Bench Pro; DeepSWE
Explains anomalous behavior across coding benchmarks
Engagement: n/a · subagent
Agentic inference hardware benchmark
Models: DeepSeek V4 Pro
Benchmarks: AA-AgentPerf
Agents per MW across Blackwell/Hopper/AMD
Engagement: n/a · subagent
AA Intelligence Index v4.1
Models: GPT-5.5; Opus 4.8; DeepSeek V4 Pro
Benchmarks: AA Intelligence Index v4.1; GDPval-AA v2
Shift toward agentic workloads/per-task metrics
Engagement: n/a · subagent
GDPval-AA v2 upgrade
Models: Claude Fable 5; GPT-5.5
Benchmarks: GDPval-AA v2
Rebaselines Elo to human performance; longer agent trajectories
Engagement: n/a · subagent
GLM-5.2 open-weight frontier
Models: GLM-5.2; MiniMax-M3; DeepSeek V4 Pro; Kimi K2.6
Benchmarks: AA Intelligence Index; CritPt; HLE
GLM-5.2 becomes leading open-weight model in AA index
Engagement: n/a · web_search exact X result
GLM-5.2 GDPval strength
Models: GLM-5.2; MiniMax-M3; DeepSeek V4 Pro; GPT-5.5
Benchmarks: GDPval-AA v2
GLM-5.2 leads open weights on agentic knowledge work
Engagement: n/a · web_search exact X result
GLM-5.2 physics reasoning
Models: GLM-5.2; GPT-5.5 Pro; Opus 4.8; DeepSeek V4 Pro
Benchmarks: CritPt
Open-weight model ties Opus 4.8 on unpublished physics problems
Engagement: n/a · subagent
GLM-5.2 real-world agent benchmark
Models: GLM-5.2; Fable 5; Opus 4.8; GPT-5.5; MiniMax-M3
Benchmarks: GDPval-AA
GLM-5.2 #3 overall and top open-weight on work tasks
Engagement: n/a · web_search exact X result
AA-Briefcase task duration
Models: Fable 5; GLM-5.2; frontier models
Benchmarks: AA-Briefcase
Shows agentic knowledge-work task duration and cost/time implications
Engagement: n/a · subagent
GLM-5.2 verbosity/cost
Models: GLM-5.2; GPT-5.5; Opus 4.8; Gemini 3.5 Flash
Benchmarks: AA Intelligence Index; AA-Omniscience; GDPval-AA
Shows high output-token use and benchmark efficiency tradeoffs
Engagement: n/a · web_search exact X result
Claude Sonnet 5 AA score
Models: Claude Sonnet 5; GPT-5.5; Opus 4.8
Benchmarks: AA Intelligence Index; AA-Briefcase; GDPval-AA
Sonnet 5 reaches strong score but with cost/token caveats
Engagement: n/a · subagent
GLM-5.2 GDPval amplification
Models: GLM-5.2; GPT-5.5
Benchmarks: GDPval-AA
High-reach amplification of open-weight model beating GPT-5.5 on GDPval-AA
Engagement: n/a · web_search exact X result
GLM-5.2 Chinese benchmark summary
Models: GLM-5.2; MiniMax-M3; DeepSeek V4 Pro; Kimi K2.6
Benchmarks: AA Intelligence Index; GDPval-AA
Chinese-language amplification of GLM-5.2 benchmark results
Engagement: n/a · web_search exact X result
Claude Sonnet 5 benchmark comparison
Models: Claude Sonnet 5; Sonnet 4.6; GPT-5.5; Gemini 3.5 Flash
Benchmarks: benchmark comparison
Model-vs-model benchmark comparison post
Engagement: n/a · web_search exact X result
AA benchmark commentary on Fable 5
Models: Claude Fable 5; GPT-5.5; Gemini 3.1 Pro
Benchmarks: AA overall intelligence benchmarks
Commentary translating AA score gaps into model-buying context
Engagement: n/a · web_search exact X result
BiomniBench launch
Models: AI agents
Benchmarks: BiomniBench
Process-level biomedical agent benchmark post
Engagement: n/a · prior search result
Artificial Analysis GPT-5.6 Coding Agent Index
Models: GPT-5.6; Claude Fable 5
Benchmarks: Coding Agent Index; DeepSWE; Terminal-Bench
Follow-up/related AA posts on GPT-5.6 coding-agent evals
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Artificial Analysis AutomationBench-AA Zapier
Models: Gemini; Claude; GPT
Benchmarks: AutomationBench-AA
SaaS-workflow agent benchmark
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Artificial Analysis Harvey LAB-AA legal benchmark
Models: Claude; GPT; GLM
Benchmarks: Harvey LAB-AA
Legal-agent benchmark and model cost spread
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Artificial Analysis AA-Briefcase
Models: Fable 5; GLM-5.2
Benchmarks: AA-Briefcase
Agentic business-work benchmark
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Artificial Analysis AA-Omniscience
Models: Fable 5; GPT-5.5; Gemini
Benchmarks: AA-Omniscience
Knowledge/hallucination benchmark
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Artificial Analysis CritPt physics benchmark
Models: GLM-5.2; Opus 4.8
Benchmarks: CritPt
Scientific reasoning benchmark
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Artificial Analysis HLE benchmark model
Models: GLM-5.2; GPT-5.5; Claude
Benchmarks: HLE
Hard reasoning benchmark commentary
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
LM Arena leaderboard GPT Claude Gemini
Models: GPT; Claude; Gemini
Benchmarks: Chatbot Arena; Arena leaderboard
Human preference/model-ranking updates
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Chatbot Arena leaderboard new model
Models: frontier models
Benchmarks: Chatbot Arena
Arena ranking update
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
LiveBench leaderboard GPT Claude Gemini
Models: GPT; Claude; Gemini
Benchmarks: LiveBench
Contamination-resistant benchmark updates
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Aider Polyglot GPT Claude benchmark
Models: GPT; Claude; Gemini
Benchmarks: Aider Polyglot
Coding model benchmark updates
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Paul Gauthier aider benchmark Claude GPT
Models: GPT; Claude
Benchmarks: Aider Polyglot
Creator commentary on coding model performance
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
SWE-bench GPT Claude benchmark
Models: GPT; Claude; Gemini
Benchmarks: SWE-bench Verified; SWE-bench Multilingual
Coding-agent benchmark updates
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Terminal-Bench GPT Claude benchmark
Models: GPT; Claude
Benchmarks: Terminal-Bench
Terminal-agent benchmark
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
METR GPT-5.6 benchmark gaming
Models: GPT-5.6 Sol
Benchmarks: METR agent evals
Agentic benchmark gaming/behavioral caveats
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Epoch AI benchmark GPT Claude Gemini
Models: frontier models
Benchmarks: benchmark trends
Macro benchmark/progress analysis
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
HELM Stanford leaderboard GPT Claude
Models: GPT; Claude; Gemini
Benchmarks: HELM; MedHELM
Holistic LLM eval framework updates
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Hugging Face Open LLM Leaderboard new model
Models: open models
Benchmarks: Open LLM Leaderboard
Open model leaderboard updates
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Nathan Lambert open model benchmark
Models: Llama; Qwen; DeepSeek; GLM
Benchmarks: open model benchmarks
Post-training/open-model evaluation commentary
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
swyx GPT-5.6 benchmark eval
Models: GPT-5.6; Claude
Benchmarks: benchmarks/evals
AI engineering interpretation of benchmark results
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Simon Willison GPT-5.6 benchmark eval
Models: GPT-5.6; Claude; Gemini
Benchmarks: hands-on evals
Practical testing of model behavior
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Hamel Husain eval benchmark model
Models: LLMs
Benchmarks: evals
Product-eval methodology and model testing
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
OpenRouter model benchmark price performance
Models: frontier/open models
Benchmarks: price/performance
Routing/API market signals
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Papers with Code leaderboard LLM benchmark
Models: LLMs
Benchmarks: leaderboards
Benchmark leaderboard amplification
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Scale Labs leaderboard model benchmarks
Models: frontier models
Benchmarks: Scale Labs leaderboards
Expert benchmark leaderboards
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Scale Labs DrugDiscoveryBench
Models: GPT; Gemini; Claude
Benchmarks: DrugDiscoveryBench
Life-science/drug discovery benchmark
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Phylo DrugDiscoveryBench benchmark
Models: AI agents
Benchmarks: DrugDiscoveryBench; BiomniBench
Biomedical agent benchmark
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
OpenAI LifeSciBench GPT-Rosalind
Models: GPT-Rosalind; GPT-5.5
Benchmarks: LifeSciBench
Life-science model benchmark
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
OpenAI HealthBench GPT-5.6
Models: GPT-5.6
Benchmarks: HealthBench
Medical model benchmark
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
OpenAI Terminal-Bench GPT-5.6 Sol
Models: GPT-5.6 Sol
Benchmarks: Terminal-Bench
Coding/terminal benchmark
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
OpenAI Developers GPT-5.6 evals
Models: GPT-5.6
Benchmarks: coding; tool use
Developer benchmark/performance details
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Anthropic Claude Sonnet 5 benchmark
Models: Claude Sonnet 5
Benchmarks: coding; reasoning
Official Claude benchmark claims
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Google DeepMind Gemini benchmark
Models: Gemini
Benchmarks: reasoning; coding; multimodal
Official Gemini performance claims
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Mistral AI model benchmark
Models: Mistral models
Benchmarks: open/frontier benchmarks
Mistral model release performance
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Meta AI Llama benchmark
Models: Llama
Benchmarks: open model benchmarks
Llama benchmark/release comparisons
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
DeepSeek model benchmark
Models: DeepSeek
Benchmarks: coding/reasoning benchmarks
DeepSeek model-performance posts
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Qwen benchmark model performance
Models: Qwen
Benchmarks: open model benchmarks
Qwen model benchmark results
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Z.ai GLM-5.2 benchmark
Models: GLM-5.2
Benchmarks: AA Index; GDPval-AA
Source account for GLM-5.2 benchmark claims
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Cursor Grok 4.5 benchmark
Models: Grok 4.5
Benchmarks: coding benchmarks
Coding-product benchmark context
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Cognition Devin benchmark
Models: coding agents
Benchmarks: SWE-bench
Coding agent performance
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Windsurf model benchmark coding
Models: coding models
Benchmarks: coding benchmark
AI IDE model-performance posts
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Replit Agent benchmark
Models: coding agents
Benchmarks: coding/task benchmarks
AI coding-agent performance
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
LangChain agent benchmark
Models: agent models
Benchmarks: agent evals
Agent framework evaluation
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
LlamaIndex RAG benchmark
Models: LLMs
Benchmarks: RAG evals
RAG/model evaluation
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Modal GPU inference benchmark
Models: inference stacks
Benchmarks: latency/throughput
Inference-performance benchmarks
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Anyscale LLM serving benchmark
Models: LLMs
Benchmarks: serving throughput
Infrastructure benchmarks
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Together AI benchmark throughput
Models: open models
Benchmarks: inference speed/cost
Model-serving benchmark posts
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Fireworks AI inference benchmark
Models: open models
Benchmarks: serving latency/throughput
Inference benchmark posts
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Groq inference benchmark
Models: LLMs
Benchmarks: tokens/sec latency
Inference speed benchmarks
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Cerebras inference benchmark
Models: LLMs
Benchmarks: tokens/sec
Inference speed benchmarks
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Perplexity model benchmark
Models: Perplexity models
Benchmarks: search/model quality
AI search model performance
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Vellum GPT-5.6 benchmarks explained
Models: GPT-5.6
Benchmarks: benchmark analysis
Practical model comparison
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Latent Space AINews GPT-5.6 benchmarks
Models: GPT-5.6
Benchmarks: benchmark roundup
Benchmark news synthesis
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Deedy Das model benchmark
Models: GPT; Claude; Gemini
Benchmarks: benchmark summaries
High-signal benchmark commentary
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
AlphaSignal benchmark model
Models: AI models
Benchmarks: research benchmarks
AI R&D benchmark amplification
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Cameron Wolfe model benchmark
Models: AI models
Benchmarks: benchmark analysis
Technical paper/model benchmark breakdowns
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Sebastian Raschka model benchmark
Models: open models
Benchmarks: LLM benchmarks
Technical model comparison
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Philipp Schmid model benchmark
Models: open models
Benchmarks: leaderboards/inference
Hugging Face benchmark/inference testing
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Thomas Wolf open model benchmark
Models: open models
Benchmarks: leaderboards
Open-source model performance
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Clement Delangue open model benchmark
Models: open models
Benchmarks: leaderboards
Open model release/benchmark amplification
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Kaggle Abhishek model benchmark
Models: ML models
Benchmarks: Kaggle/evals
Applied model-performance tests
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Valence Labs Polaris benchmark
Models: drug discovery models
Benchmarks: Polaris
Drug-discovery benchmark platform
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Pat Walters ADMET benchmark
Models: cheminformatics models
Benchmarks: ADMET/blind challenge
Drug-discovery benchmark critique
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
TDC benchmark model
Models: therapeutic ML models
Benchmarks: TDC
Therapeutics benchmark ecosystem
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
OpenBioML benchmark
Models: biology/chem models
Benchmarks: BioML/Chem benchmarks
Open biology ML benchmark amplification
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Chai Discovery benchmark
Models: structure models
Benchmarks: protein/structure benchmarks
Biomolecular model performance
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Rosetta Commons benchmark
Models: protein design models
Benchmarks: protein modeling benchmarks
Protein modeling benchmark network
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
MedARC Medmarks benchmark
Models: medical LLMs
Benchmarks: Medmarks
Medical LLM benchmark
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
AI Security Institute Inspect Evals
Models: frontier models
Benchmarks: Inspect Evals
Model safety/eval framework
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
CAIS WMDP benchmark
Models: LLMs
Benchmarks: WMDP
Bio/chem/cyber risk benchmark
Engagement: n/a · X search URL because exact post URL was not surfaced unauthenticated
Kept as a separate domain-specific watchlist for DDDBench/MMAI/BioML-style outreach.
| # | Name | X handle | Followers | Technical | Performance focus | Why / rationale |
|---|---|---|---|---|---|---|
| 1 | OpenAI | @OpenAI | very large / n.a. | Technical | LifeSciBench; GeneBench-Pro; life-science model evaluation | LifeSciBench: 750 expert-authored tasks, 173 scientist contributors, 19,020 rubric criteria. Direct benchmark comparator and high-impact launch/amplification target. |
| 2 | Tacit Labs | @tacitlabsco | n/a | Technical | LifeSciBench partner / biology verification loop | Referenced in LifeSciBench launch ecosystem with OpenAI. Potential partner/peer for life-science task design and validation. |
| 3 | Nina Fitzpatrick | @ninklefitz | n/a | Technical | LifeSciBench / Tacit Labs | Named in LifeSciBench/Tacit Labs launch amplification. Likely useful individual contact around biology eval design. |
| 4 | Andrew Droste | @AmDroste | n/a | Technical | LifeSciBench / Tacit Labs | Named in LifeSciBench/Tacit Labs launch amplification. Likely useful individual contact around life-science verification. |
| 5 | FutureHouse | @FutureHouseSF | n/a | Technical | Biology/science agents, LAB-Bench, scientific task evaluation | LAB-Bench and FutureHouse scientific-agent ecosystem; BixBench repo under Future-House. Top target for scientific-agent benchmark comparison/amplification. |
| 6 | Phylo | @phylo_bio | n/a | Technical | Biomni, BiomniBench, DrugDiscoveryBench, biomedical agents | DrugDiscoveryBench with Scale Labs; BiomniBench process-level evaluation. Direct DDDBench peer/benchmark partner and high-priority outreach target. |
| 7 | Scale Labs | @ScaleAILabs | n/a | Technical | DrugDiscoveryBench and agent evals | DrugDiscoveryBench: 82 early-stage DD tasks with expert rubrics and leaderboard. Direct benchmark comparison; useful for model-vs-harness narrative. |
| 8 | Scale AI | @scale_AI | large / n.a. | Technical | AI data/evals; Scale Labs amplification | Scale blog and labs leaderboard for DrugDiscoveryBench. Enterprise-eval distribution channel. |
| 9 | Kexin Huang | @KexinHuang5 | n/a | Technical | Biomni, BiomniBench, AI agents for biomedical research | Posted BiomniBench launch: process-level evaluation of agents on long-horizon biology tasks. Strong KOL for agentic biomedical benchmarks and scientific-agent systems. |
| 10 | Therapeutics Data Commons | @ProjectTDC | ~1.3K | Technical | Drug-discovery ML datasets/tasks/leaderboards | TDC platform for AI-ready therapeutic datasets and leaderboards. Foundational benchmark ecosystem for DDDBench positioning. |
| 11 | Jure Leskovec | @jure | ~44.7K | Technical | TDC, graph ML, biomedical ML | TDC / Harvard-Stanford biomedical ML ecosystem. Senior benchmark and graph-ML amplifier. |
| 12 | Valence Labs | @valence_ai | ~5.9K | Technical | Polaris drug-discovery benchmark platform | PolarisHub benchmark/data platform for drug discovery. Highly relevant for industry-standard DD benchmarking. |
| 13 | Pat Walters | @wpwalters | ~6.3K | Technical | Cheminformatics benchmark critique, ADMET, blind challenges | OpenADMET/blind-challenge and benchmark-validity commentary. High-credibility evaluator/KOL for DD benchmark quality. |
| 14 | OpenBioML | @openbioml | ~4.1K | Technical | Open biology ML, ChemNLP, model-risk/eval discussion | Community journal clubs and BioML paper amplification. Community amplifier for open BioML benchmarking. |
| 15 | Biology+AI Daily | @BiologyAIDaily | n/a | Technical | AI-bio paper/benchmark amplification | Amplified BixBench and other AI-biology benchmark papers. Useful awareness channel for DDDBench/MMAI benchmark posts. |
| 16 | Stephen Turner | @strnr | ~31.3K | Technical | Genomics, computational biology, AIxBio commentary | Genomics and biosecurity/AI benchmark commentary. Strong commentator/amplifier for AI-bio benchmark credibility. |
| 17 | Jablonka Lab | @jablonkagroup | ~359 | Technical | ChemBench, chemistry/materials LLM evaluation | ChemBench benchmark for chemistry reasoning and tool use. Relevant analog for chemistry/LLM eval methodology. |
| 18 | Andrew White | @andrewwhite01 | n/a | Technical | ChemCrow, PaperQA, scientific agents, chemistry LLMs | ChemCrow and scientific-agent evaluation work. Very strong chemistry/science-agent benchmark KOL. |
| 19 | Philippe Schwaller | @pschwllr | n/a | Technical | Chemical language models, RXN, ChemCrow ecosystem | ChemCrow / reaction prediction / chemistry AI work. Good benchmark reviewer and chemistry-agent evaluator. |
| 20 | Chai Discovery | @chaidiscovery | ~5.6K | Technical | Biomolecular structure prediction benchmarks | Chai-1, chai-lab, structure prediction comparisons. Protein/structure benchmark comparison channel. |
| 21 | Google DeepMind | @GoogleDeepMind | ~1.2M | Technical | AlphaFold/CASP, Gemini science AI | AlphaFold and Isomorphic science-AI narratives. Huge amplifier; benchmark framing authority. |
| 22 | Rosetta Commons | @RosettaCommons | ~1.9K | Technical | Protein modeling/design benchmarks and community tools | Rosetta, CASP/CAPRI-related modeling ecosystem. Protein-design benchmark credibility network. |
| 23 | ChEMBL Database | @ChEMBL | ~2.9K | Technical | Drug-discovery data resource used in benchmarks | ChEMBL data/resource updates. Useful data/benchmark community account. |
| 24 | Papers with Code | @paperswithcode | ~114.7K | Technical | ML benchmark/leaderboard amplification | Benchmark and paper leaderboard distribution. Broad benchmark visibility channel. |
| 25 | Hugging Face | @huggingface | ~689K | Technical | Open leaderboards and model hubs | Open Medical LLM Leaderboard; dataset/model benchmark hosting. High-impact hosting/amplification route for DDDBench/MMAI. |
| 26 | Percy Liang / Stanford HELM | @percyliang | ~106.5K | Technical | HELM, MedHELM, holistic LLM evals | Stanford CRFM HELM and medical evaluation frameworks. Credible LLM-eval benchmark KOL. |
| 27 | METR | @METR_Evals | ~25.2K | Technical | Agent/R&D capability evals, time-horizon evals | Autonomous capability measurement and eval methodology. Useful for agent-eval methodology framing. |
| 28 | UK AI Security Institute | @AISecurityInst | ~7.6K | Technical | Inspect Evals; chem/bio/frontier model evals | Inspect eval framework and frontier-model safety evaluations. Strong eval-framework/governance target. |
| 29 | Center for AI Safety | @CAIS / @ai_risks | ~8.2K | Technical | WMDP bio/chem/cyber hazardous-knowledge benchmark | WMDP benchmark and biorisk/chem risk evals. Relevant to bio/chem LLM evals; outreach should be careful and risk-aware. |
| 30 | Epoch AI | @EpochAIResearch | ~45.7K | Technical | Benchmark datasets/trends and biorisk eval analysis | Epoch benchmark tracking and AI trend analysis. Methodology/benchmark-analysis amplifier. |
| 31 | MLCommons | @MLCommons | ~3.6K | Technical | Medical AI, safety, and science benchmark working groups | MLCommons medical/data working groups. Standards-setting outreach target. |
| 32 | Tanishq M. Abraham / MedARC | @iScienceLuvr | n/a | Technical | Medmarks medical LLM benchmark suite | Medmarks benchmark posts and GitHub project. Strong medical LLM benchmark KOL. |
| # | Name | X handle | Followers | Technical | Performance focus | Why / rationale |
|---|---|---|---|---|---|---|
| 1 | Artificial Analysis | @ArtificialAnlys | ~25K | Technical | frontier benchmarks; cost/speed/quality; coding agents | Independent model quality, intelligence, speed, price, cost-per-task, and coding-agent comparisons. Independent model quality, intelligence, speed, price, cost-per-task, and coding-agent comparisons. |
| 2 | LM Arena | @lmarena_ai | ~80K | Technical | frontier benchmarks; human preference arena | Publishes Chatbot Arena / model leaderboard results and arena methodology. Publishes Chatbot Arena / model leaderboard results and arena methodology. |
| 3 | LMSYS Org | @lmsysorg | ~60K | Technical | frontier benchmarks; arena methodology | Originator of Chatbot Arena/FastChat-style model evaluation work. Originator of Chatbot Arena/FastChat-style model evaluation work. |
| 4 | LiveBench | @livebench_ai | ~10K | Technical | contamination-resistant LLM benchmark | Maintains live/refreshing benchmark results to reduce contamination and stale evals. Maintains live/refreshing benchmark results to reduce contamination and stale evals. |
| 5 | Aider | @aider_chat | ~30K | Technical | Aider Polyglot; coding model performance | Publishes practical coding model benchmark results from Aider workflows. Publishes practical coding model benchmark results from Aider workflows. |
| 6 | Paul Gauthier | @paulgauthier | ~20K | Technical | coding model performance | Creator of Aider; frequent publisher/interpreter of coding model leaderboard results. Creator of Aider; frequent publisher/interpreter of coding model leaderboard results. |
| 7 | SWE-bench | @SWEbench | ~20K | Technical | coding agents; SWE-bench Verified/Multilingual | Official software-engineering benchmark and leaderboard for AI coding agents. Official software-engineering benchmark and leaderboard for AI coding agents. |
| 8 | OpenAI Developers | @OpenAIDevs | n/a | Technical | API/model performance and coding/tool-use updates | Developer-facing model release and performance details. Developer-facing model release and performance details. |
| 9 | Vals AI | @ValsAI | n/a | Technical | frontier model benchmarks across domains | Publishes model rankings across legal, financial, coding, multimodal, and other private benchmarks. Publishes model rankings across legal, financial, coding, multimodal, and other private benchmarks. |
| 10 | Andrej Karpathy | @karpathy | ~3.2M | Technical | AI researcher / educator | One of the highest signal technical AI explainers on X. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 11 | Andrew Ng | @AndrewYNg | ~1.5–1.6M | Technical | AI education / applied ML | Major applied-AI educator and practitioner influence. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 12 | Demis Hassabis | @demishassabis | ~1.3M | Technical | Google DeepMind CEO | High-authority AI-for-science and AGI lab voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 13 | Yann LeCun | @ylecun | ~0.94–1.2M | Technical | Meta AI / Turing laureate | Major technical counterweight to LLM-scaling consensus. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 14 | Ilya Sutskever | @ilyasut | ~760.9K | Technical | SSI / frontier research | Rare posts carry frontier-lab/research weight. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 15 | Aravind Srinivas | @AravSrinivas | ~800.6K | Semi-technical | Perplexity CEO | AI search and consumer-AI distribution leader. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 16 | Fei-Fei Li | @drfeifei | ~632–781K | Technical | World Labs / Stanford | Major AI research and human-centered AI authority. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 17 | François Chollet | @fchollet | ~566–704K | Technical | Keras / ARC-AGI | Important benchmark/reasoning voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 18 | Geoffrey Hinton | @geoffreyhinton | ~489–618K | Technical | Deep learning pioneer | High-authority voice on deep learning and AI risk. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 19 | Dario Amodei | @DarioAmodei | ~478.6K | Semi-technical | Anthropic CEO | Claude/frontier AI strategy and safety framing. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 20 | Jeff Dean | @JeffDean | ~361–448K | Technical | Google DeepMind / Google Research | Elite systems + ML research credibility. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 21 | Clément Delangue | @ClementDelangue | ~452.1K | Semi-technical | Hugging Face CEO | Open-source AI distribution and community leader. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 22 | Alexandr Wang | @alexandr_wang | ~541.7K | Semi-technical | Meta AI / Scale founder | Data, evals, and enterprise AI influence. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 23 | Jim Fan | @DrJimFan | ~435–486K | Technical | NVIDIA robotics / agents | High-signal physical AGI and robotics commentator. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 24 | Sebastian Raschka | @rasbt | ~476K | Technical | ML/LLM educator | Hands-on LLM training and architecture education. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 25 | Soumith Chintala | @soumithchintala | ~247–311K | Technical | PyTorch / Meta | PyTorch ecosystem credibility. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 26 | Lilian Weng | @lilianweng | ~88–267K | Technical | AI research / agents / safety | Widely cited technical explainer. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 27 | Jeremy Howard | @jeremyphoward | ~236–289K | Technical | fast.ai / Answer.AI | Applied technical AI education and open models. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 28 | Emad Mostaque | @EMostaque | ~328.3K | Semi-technical | Stability AI founder | Open-source AI and generative media commentary. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 29 | Logan Kilpatrick | @OfficialLoganK | ~339.2K | Semi-technical | Google AI Studio / devrel | Developer-facing AI launch signal. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 30 | Simon Willison | @simonw | ~154.5K | Technical | Developer / AI tooling analyst | Top practical technical evaluator of LLM tools. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 31 | Hamel Husain | @HamelHusain | n/a | Technical | AI evals / applied LLM systems | High-signal practitioner voice on AI evals. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 32 | Aidan McLaughlin | @aidan_mclau | ~47.8K | Technical | OpenAI research | Useful model-design and frontier-lab research signal. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 33 | Ian Goodfellow | @goodfellow_ian | ~315.7K | Technical | Deep learning researcher | Foundational ML researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 34 | Oriol Vinyals | @oriolvinyalsml | ~182.4K | Technical | Google DeepMind research | Major DeepMind technical leader. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 35 | Nando de Freitas | @NandoDF | ~108.9K | Technical | Microsoft AI / ex-DeepMind | High-credibility research voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 36 | Abhishek Thakur | @abhi1thakur | ~83.1K | Technical | Kaggle / Hugging Face AutoTrain | Applied ML competition and tooling expert. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 37 | Gary Marcus | @GaryMarcus | ~213.5K | Semi-technical | AI critic / cognitive scientist | Influential skeptical AI commentator. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 38 | Data Chaz | @DataChaz | ~114.1K | Semi-technical | Developer advocate / Streamlit | Practical AI/data app educator. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 39 | Jürgen Schmidhuber | @SchmidhuberAI | ~98.1K | Technical | AI researcher | Longstanding ML research figure. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 40 | Pedro Domingos | @pmddomingos | ~74.3K | Technical | ML professor / author | Recognized ML academic voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 41 | Santiago Valdarrama | @svpino | ~383.3K | Technical | ML educator | Large practical ML learning audience. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 42 | Bojan Tunguz | @tunguz | ~251.1K | Technical | ML engineer / Kaggle | Applied ML/Kaggle KOL. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 43 | Akshay Pachaar | @akshay_pachaar | ~240K | Semi-technical | LLM/agents educator | Large applied AI education audience. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 44 | Hugo Larochelle | @hugo_larochelle | ~115.9K | Technical | Google DeepMind researcher | High-quality ML research voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 45 | Peyman Milanfar | @docmilanfar | ~100.1K | Technical | Google computational imaging | Technical computer-vision KOL. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 46 | Dan Kornas | @dankornas | ~87.1K | Technical | ML engineer / educator | Practical ML engineering audience. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 47 | Kevin Patrick Murphy | @sirbayes | ~64.7K | Technical | Google DeepMind researcher | Authoritative ML textbook/research voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 48 | Gautam Kamath | @thegautamkamath | ~59.7K | Technical | ML professor / Vector Institute | Technical ML theory voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 49 | Sara Hooker | @sarahookr | ~53.3K | Technical | Cohere Labs / model efficiency | Important efficient/open AI researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 50 | Durk Kingma | @dpkingma | ~51.3K | Technical | Anthropic / VAE/Adam inventor | Foundational ML researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 51 | Andriy Burkov | @burkov | ~50.8K | Technical | ML author | Strong ML educational voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 52 | Jean de Nyandwi | @Jeande_d | ~46.9K | Technical | CMU / NLP evals | Rising technical AI research voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 53 | Shakir Mohamed | @shakir_za | ~41.7K | Technical | DeepMind researcher | Technical DeepMind/social-good voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 54 | Ferenc Huszár | @fhuszar | ~37.9K | Technical | ML professor | Technical ML/statistics KOL. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 55 | Andrew Gordon Wilson | @andrewgwils | ~35.5K | Technical | ML professor | Technical ML academic. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 56 | Ben Recht | @beenwrekt | ~32.9K | Technical | Berkeley ML/control | Sharp technical ML critic. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 57 | Zoubin Ghahramani | @ZoubinGhahrama1 | ~32.6K | Technical | Google DeepMind / Cambridge | Senior ML research authority. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 58 | Jakob Foerster | @j_foerst | ~22.7K | Technical | Oxford/Meta ML researcher | Technical RL/agents researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 59 | Nils Reimers | @Nils_Reimers | ~14.6K | Technical | Cohere / SBERT | Important technical retrieval/embedding voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 60 | Nathan Raw | @_nateraw | ~9.5K | Technical | ML hacker / Hugging Face alumnus | Hands-on ML builder. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 61 | Alicia Curth | @AliciaCurth | ~4.7K | Technical | ML/statistics researcher | Technical ML/statistics voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 62 | Yoshua Bengio | @yoshuabengio | n/a | Technical | Deep learning pioneer | Turing laureate and major AI safety/research voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 63 | Chris Olah | @chrisolah | n/a | Technical | Anthropic / interpretability | Core interpretability KOL. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 64 | Anca Dragan | @ancadragan | n/a | Technical | Berkeley robotics / HRI | Technical robotics/HRI authority. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 65 | Pieter Abbeel | @pabbeel | n/a | Technical | Berkeley robotics / Covariant | Major robotics/RL KOL. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 66 | Chelsea Finn | @chelseabfinn | n/a | Technical | Stanford robotics / meta-learning | High-signal robotics/learning researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 67 | Sergey Levine | @svlevine | n/a | Technical | Berkeley robotics / RL | Top technical robotics/RL researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 68 | Been Kim | @beenkim | n/a | Technical | Google interpretability | Important technical interpretability voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 69 | Samy Bengio | @samybengio | n/a | Technical | Apple / ex-Google Brain | Senior deep learning researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 70 | Kyunghyun Cho | @kchon2020 | n/a | Technical | NYU / NLP | Technical NLP researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 71 | Max Welling | @wellingmax | n/a | Technical | CuspAI / ML professor | Technical ML/science-AI voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 72 | David Ha | @hardmaru | n/a | Technical | AI researcher / generative systems | Technical and creative AI research voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 73 | Alex Graves | @alexjgraves | n/a | Technical | DeepMind researcher | Foundational sequence-model researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 74 | Quoc Le | @quocleix | n/a | Technical | Google Brain / AutoML | Senior technical Google AI researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 75 | Leslie Kaelbling | @lpkaelbling | n/a | Technical | MIT AI/robotics | Senior AI/robotics authority. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 76 | Aidan Gomez | @aidangomez | n/a | Technical | Cohere CEO / Transformer coauthor | Technical founder with model lineage. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 77 | Rachel Thomas | @math_rachel | n/a | Technical | fast.ai / AI education | Technical educator and fast.ai cofounder. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 78 | Thomas Wolf | @Thom_Wolf | n/a | Technical | Hugging Face cofounder | Core open-source AI platform builder. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 79 | Julien Chaumond | @julien_c | n/a | Technical | Hugging Face CTO/cofounder | Important open-source AI infra voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 80 | Sebastian Thrun | @SebastianThrun | n/a | Technical | Udacity / robotics | Robotics/AI education pioneer. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 81 | Abeba Birhane | @Abebab | n/a | Semi-technical | AI accountability researcher | Strong responsible-AI researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 82 | Rediet Abebe | @red_abebe | n/a | Technical | AI for social good | Technical/public-interest AI voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 83 | Melanie Mitchell | @MelMitchell1 | n/a | Technical | Complexity / AI researcher | Technical and skeptical AI researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 84 | Aurélien Géron | @aureliengeron | n/a | Technical | ML author | Popular practical ML educator. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 85 | Chip Huyen | @chipro | n/a | Technical | AI engineering author | High-signal AI engineering voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 86 | Monica Rogati | @mrogati | n/a | Technical | Data science leader | Technical data/AI product KOL. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 87 | Inioluwa Deborah Raji | @rajiinio | n/a | Semi-technical | AI accountability researcher | Important AI auditing voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 88 | Arvind Narayanan | @random_walker | n/a | Technical | Princeton AI/society | Technical AI and policy critic. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 89 | Vincent Warmerdam | @fishnets88 | n/a | Technical | Calmcode / ML tooling | High-signal practical technical educator. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 90 | Sasha Luccioni | @SashaMTL | n/a | Semi-technical | Hugging Face / AI climate | Technical responsible-AI voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 91 | Jan Leike | @janleike | n/a | Technical | Anthropic / alignment | Top technical alignment researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 92 | Paul Christiano | @paulc443 | n/a | Technical | Alignment Research Center | Core technical alignment voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 93 | Percy Liang | @percyliang | ~106.5K | Technical | Stanford NLP / HELM | Top LLM-evaluation researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 94 | Jacob Andreas | @jacobandreas | n/a | Technical | MIT NLP | Technical NLP/agents researcher. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 95 | Dan Hendrycks | @DanHendrycks | n/a | Technical | Center for AI Safety | Important eval/safety benchmark creator. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 96 | Scott Aaronson | @scottaaronson | n/a | Technical | Theoretical CS / OpenAI alumnus | Technical theory/safety voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 97 | Swyx | @swyx | n/a | Technical | AI engineering / Latent Space | Top AI engineering community builder. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 98 | Deedy Das | @deedydas | n/a | Semi-technical | AI investor / curator | High-signal AI startup/model curator. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 99 | Linus Lee | @thesephist | n/a | Technical | Notion AI researcher | Technical AI product/research voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 100 | Nathan Labenz | @labenz | n/a | Semi-technical | Cognitive Revolution podcast | Strong AI interview/eval commentator. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 101 | Nathan Lambert | @natolambert | n/a | Technical | AI2 / Interconnects | Top technical post-training/open-model voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 102 | arXiv | @arxiv_org | n/a | Technical | Research preprint platform | Raw research discovery channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 103 | AlphaXiv | @alphaxiv | n/a | Technical | Paper discovery platform | AI paper discovery/amplification. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 104 | Cameron Wolfe | @cwolferesearch | n/a | Technical | AI researcher / writer | High-signal technical AI explainer. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 105 | Vikhyat K. | @vikhyatk | n/a | Technical | Moondream / small VLMs | Practical model builder. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 106 | OpenAI | @OpenAI | n/a | Semi-technical | Frontier AI lab | Primary OpenAI launch channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 107 | Anthropic | @AnthropicAI | n/a | Semi-technical | Frontier AI lab | Primary Anthropic launch channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 108 | Google DeepMind | @GoogleDeepMind | ~1.2M | Technical | Frontier AI/science lab | Major technical AI lab channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 109 | Meta AI | @MetaAI | n/a | Technical | Open models / AI lab | Key open-model lab channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 110 | Mistral AI | @MistralAI | n/a | Technical | Frontier/open model lab | European frontier model channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 111 | Hugging Face | @huggingface | ~689K | Semi-technical | Open-source AI platform | High-impact open AI distribution. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 112 | Cohere | @cohere | n/a | Semi-technical | Enterprise LLM lab | Enterprise LLM platform channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 113 | Together AI | @togethercompute | n/a | Technical | Open model infra | Open AI infrastructure channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 114 | xAI | @xai | n/a | Semi-technical | Frontier AI lab | xAI primary launch channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 115 | OpenRouter | @OpenRouterAI | n/a | Semi-technical | Model routing/API platform | Useful model market/API signal. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 116 | Replicate | @replicate | n/a | Technical | Model hosting/API | Model deployment ecosystem channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 117 | Weights & Biases | @weights_biases | n/a | Semi-technical | MLOps / evals | MLOps and AI eval community. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 118 | LangChain | @LangChainAI | n/a | Technical | LLM app framework | Key AI agent/RAG developer channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 119 | LlamaIndex | @llama_index | n/a | Technical | LLM data/RAG framework | Important AI engineering framework. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 120 | Modal | @modal_labs | n/a | Technical | AI compute platform | Technical AI infra channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 121 | Anyscale | @AnyscaleCompute | n/a | Technical | Ray / AI infra | Important scalable AI systems channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 122 | Scale AI | @scale_AI | n/a | Semi-technical | Data/evals platform | Evals/data infrastructure influence. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 123 | Scale Labs | @ScaleAILabs | n/a | Technical | AI evals / benchmarks | High-signal eval/benchmark channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 124 | METR | @METR_Evals | ~25.2K | Technical | AI capability evals | Important agent-eval methodology voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 125 | UK AI Security Institute | @AISecurityInst | ~7.6K | Semi-technical | Model safety/evals | Government eval-framework voice. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 126 | Center for AI Safety | @ai_risks | ~8.2K | Semi-technical | AI safety benchmarks | Bio/chem/cyber eval relevance. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 127 | Epoch AI | @EpochAIResearch | ~45.7K | Semi-technical | AI trends/benchmarks | Methodology/benchmark analysis amplifier. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 128 | MLCommons | @MLCommons | ~3.6K | Semi-technical | AI standards/benchmarks | Standards-setting benchmark channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 129 | Papers with Code | @paperswithcode | ~114.7K | Semi-technical | Benchmark leaderboard platform | Broad benchmark visibility channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 130 | Therapeutics Data Commons | @ProjectTDC | ~1.3K | Technical | Drug-discovery ML benchmarks | Foundational drug-discovery benchmark ecosystem. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 131 | Phylo | @phylo_bio | n/a | Technical | Biomedical agents / DrugDiscoveryBench | Direct DDDBench peer/partner target. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 132 | Kexin Huang | @KexinHuang5 | n/a | Technical | Biomedical AI agents | High-signal biomedical-agent benchmark KOL. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 133 | FutureHouse | @FutureHouseSF | n/a | Technical | Scientific agents / LAB-Bench | Top scientific-agent benchmark target. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 134 | Valence Labs | @valence_ai | ~5.9K | Technical | Drug-discovery benchmarks | Relevant industry-standard DD benchmarking channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 135 | Pat Walters | @wpwalters | ~6.3K | Technical | Cheminformatics / benchmark critique | High-credibility DD benchmark evaluator. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 136 | OpenBioML | @openbioml | ~4.1K | Technical | Open biology ML | Community amplifier for open BioML benchmarks. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 137 | Stephen Turner | @strnr | ~31.3K | Technical | Genomics / AIxBio commentary | Strong AI-bio benchmark amplifier. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 138 | Jablonka Lab | @jablonkagroup | ~359 | Technical | ChemBench / chemistry AI | Relevant chemistry benchmark group. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 139 | Andrew White | @andrewwhite01 | n/a | Technical | ChemCrow / scientific agents | Strong chemistry/science-agent benchmark KOL. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 140 | Philippe Schwaller | @pschwllr | n/a | Technical | Chemical language models | Chemistry model/eval KOL. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 141 | Chai Discovery | @chaidiscovery | ~5.6K | Technical | Structure prediction | Protein/structure benchmark channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 142 | Rosetta Commons | @RosettaCommons | ~1.9K | Technical | Protein modeling/design | Protein-design benchmark credibility network. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 143 | ChEMBL Database | @ChEMBL | ~2.9K | Technical | Drug-discovery data | Benchmark/data community account. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 144 | Tanishq Abraham / MedARC | @iScienceLuvr | n/a | Technical | Medical LLM benchmarks | Medical LLM benchmark KOL. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 145 | Stability AI | @StabilityAI | n/a | Semi-technical | Open generative AI lab | Generative AI model distribution channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 146 | Midjourney | @midjourney | n/a | Non-technical / media-business | Generative image models | Major creative AI model channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 147 | Runway | @runwayml | n/a | Semi-technical | Generative video AI | Major generative video model channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 148 | Luma AI | @LumaLabsAI | n/a | Semi-technical | Generative video/3D | Creative model product channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 149 | Cursor | @cursor_ai | n/a | Technical | AI coding IDE | Major AI coding tool channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| 150 | Cognition | @cognition_labs | n/a | Technical | AI software agents | Important coding-agent benchmark/product channel. Focused on model capability, benchmarks/evals, open-model comparisons, AI engineering tests, coding/agent performance, or model infrastructure — not general celebrity influence. |
| # | Author | Post / search URL | Date | Topic | Models | Benchmarks | Why it matters | Confidence |
|---|---|---|---|---|---|---|---|---|
| 1 | Artificial Analysis | @ArtificialAnlys | 1 day ago | GPT-5.6 benchmark launch analysis | GPT-5.6 Sol/Terra/Luna; Claude Fable 5 | Coding Agent Index; DeepSWE; Terminal-Bench v2; SWE-Atlas-QnA | Sol leads coding-agent evals and approaches Fable intelligence at lower cost | High |
| 2 | OpenAI | @OpenAI | 1 day ago | GPT-5.6 public release benchmark claims | GPT-5.6 family | Agents Last Exam; Terminal-Bench; HealthBench; AA Index | Primary model-release post for latest OpenAI benchmark family | Medium |
| 3 | OpenAI Developers | @OpenAIDevs | 1 day ago | Developer/API performance for GPT-5.6 | GPT-5.6 family | coding; tool use; context/pricing | Builder-facing model-performance details | Medium |
| 4 | Artificial Analysis | @ArtificialAnlys | 2 days ago | Grok 4.5 independent benchmark | Grok 4.5; GPT-5.5; Fable 5; Opus 4.8 | AA Intelligence Index; Coding Agent Index | Places Grok 4.5 high overall and efficient for coding agents | High |
| 5 | OpenAI | @OpenAI | 2 weeks ago | GPT-5.6 Sol Terminal-Bench SOTA | GPT-5.6 Sol | Terminal-Bench 2.1 | OpenAI says Sol sets SOTA on complex command-line workflows | High |
| 6 | Artificial Analysis | @ArtificialAnlys | 4 days ago | Capability indices by domain | Fable 5; Opus 4.8; GPT-5.5; Gemini 3.5 Flash; GLM-5.2 | AA Capability Indices | Domain-specific model ranking shifts and open-weight leaders | High |
| 7 | Artificial Analysis | @ArtificialAnlys | 4 days ago | Cost-per-task frontier comparison | DeepSeek V4 Flash; GLM-5.2; frontier models | AA Strategy & Ops Index; cost/task | Highlights cheap open-weight task completion vs frontier premium | High |
| 8 | Scale AI Labs | @ScaleAILabs | 1 week ago | DrugDiscoveryBench launch | GPT-5.5; Gemini 3.5 Flash; Opus 4.8 | DrugDiscoveryBench | Independent benchmark for early drug-discovery computational work | High |
| 9 | Wes Roth | @WesRoth | 1 week ago | DrugDiscoveryBench summary | leading AI agents | DrugDiscoveryBench | Broad amplification of life-science model benchmark | High |
| 10 | OpenAI | @OpenAI | 3 weeks ago | LifeSciBench release | GPT-Rosalind; GPT-5.5 | LifeSciBench | Life-science benchmark: GPT-Rosalind improves reported performance | High |
| 11 | Artificial Analysis | @ArtificialAnlys | 1 month ago | Claude Fable 5 tops AA Intelligence Index | Claude Fable 5; Opus 4.8 | AA Intelligence Index | Fable 5 score/rank post from independent benchmark source | High |
| 12 | Artificial Analysis | @ArtificialAnlys | 1 month ago | Fable 5 knowledge/hallucination eval | Claude Fable 5; Opus 4.8 | AA-Omniscience | Knowledge reliability / hallucination-rate comparison | High |
| 13 | Artificial Analysis | @ArtificialAnlys | 3 weeks ago | Fable 5 benchmark cost | Claude Fable 5; Opus 4.8; GPT-5.5 | AA Intelligence Index; cost/task | Shows benchmark-run cost and premium frontier pricing | High |
| 14 | Artificial Analysis | @ArtificialAnlys | 1 month ago | DeepSWE replaces SWE-Bench Pro | GPT-5.5 Codex; Claude Code Fable 5; Opus 4.8 | DeepSWE; SWE-Bench Pro | Coding-agent benchmark contamination/gameability discussion | High |
| 15 | Artificial Analysis | @ArtificialAnlys | 1 month ago | SWE-Bench Pro anomaly | GPT-5.5 Codex; Opus 4.8 | SWE-Bench Pro; DeepSWE | Explains anomalous behavior across coding benchmarks | High |
| 16 | Artificial Analysis | @ArtificialAnlys | 1 month ago | Agentic inference hardware benchmark | DeepSeek V4 Pro | AA-AgentPerf | Agents per MW across Blackwell/Hopper/AMD | High |
| 17 | Artificial Analysis | @ArtificialAnlys | 4 weeks ago | AA Intelligence Index v4.1 | GPT-5.5; Opus 4.8; DeepSeek V4 Pro | AA Intelligence Index v4.1; GDPval-AA v2 | Shift toward agentic workloads/per-task metrics | High |
| 18 | Artificial Analysis | @ArtificialAnlys | 4 weeks ago | GDPval-AA v2 upgrade | Claude Fable 5; GPT-5.5 | GDPval-AA v2 | Rebaselines Elo to human performance; longer agent trajectories | High |
| 19 | Artificial Analysis | @ArtificialAnlys | 3 weeks ago | GLM-5.2 open-weight frontier | GLM-5.2; MiniMax-M3; DeepSeek V4 Pro; Kimi K2.6 | AA Intelligence Index; CritPt; HLE | GLM-5.2 becomes leading open-weight model in AA index | High |
| 20 | Artificial Analysis | @ArtificialAnlys | 3 weeks ago | GLM-5.2 GDPval strength | GLM-5.2; MiniMax-M3; DeepSeek V4 Pro; GPT-5.5 | GDPval-AA v2 | GLM-5.2 leads open weights on agentic knowledge work | High |
| 21 | Artificial Analysis | @ArtificialAnlys | 3 weeks ago | GLM-5.2 physics reasoning | GLM-5.2; GPT-5.5 Pro; Opus 4.8; DeepSeek V4 Pro | CritPt | Open-weight model ties Opus 4.8 on unpublished physics problems | High |
| 22 | Artificial Analysis | @ArtificialAnlys | 3 weeks ago | GLM-5.2 real-world agent benchmark | GLM-5.2; Fable 5; Opus 4.8; GPT-5.5; MiniMax-M3 | GDPval-AA | GLM-5.2 #3 overall and top open-weight on work tasks | High |
| 23 | Artificial Analysis | @ArtificialAnlys | 2 weeks ago | AA-Briefcase task duration | Fable 5; GLM-5.2; frontier models | AA-Briefcase | Shows agentic knowledge-work task duration and cost/time implications | High |
| 24 | Artificial Analysis | @ArtificialAnlys | 1 week ago | GLM-5.2 verbosity/cost | GLM-5.2; GPT-5.5; Opus 4.8; Gemini 3.5 Flash | AA Intelligence Index; AA-Omniscience; GDPval-AA | Shows high output-token use and benchmark efficiency tradeoffs | High |
| 25 | Artificial Analysis | @ArtificialAnlys | 1 week ago | Claude Sonnet 5 AA score | Claude Sonnet 5; GPT-5.5; Opus 4.8 | AA Intelligence Index; AA-Briefcase; GDPval-AA | Sonnet 5 reaches strong score but with cost/token caveats | High |
| 26 | Chubby / Kimmonismus | @kimmonismus | 3 weeks ago | GLM-5.2 GDPval amplification | GLM-5.2; GPT-5.5 | GDPval-AA | High-reach amplification of open-weight model beating GPT-5.5 on GDPval-AA | Medium |
| 27 | 0xLogicrw | @0xLogicrw | 3 weeks ago | GLM-5.2 Chinese benchmark summary | GLM-5.2; MiniMax-M3; DeepSeek V4 Pro; Kimi K2.6 | AA Intelligence Index; GDPval-AA | Chinese-language amplification of GLM-5.2 benchmark results | Medium |
| 28 | AiBattle | @AiBattle_ | 1 week ago | Claude Sonnet 5 benchmark comparison | Claude Sonnet 5; Sonnet 4.6; GPT-5.5; Gemini 3.5 Flash | benchmark comparison | Model-vs-model benchmark comparison post | Medium |
| 29 | David Hendrickson | @TeksEdge | 1 month ago | AA benchmark commentary on Fable 5 | Claude Fable 5; GPT-5.5; Gemini 3.1 Pro | AA overall intelligence benchmarks | Commentary translating AA score gaps into model-buying context | Medium |
| 30 | Kexin Huang | @KexinHuang5 | within month-ish | BiomniBench launch | AI agents | BiomniBench | Process-level biomedical agent benchmark post | High |
| 31 | ArtificialAnlys | @ArtificialAnlys | past month search result | Artificial Analysis GPT-5.6 Coding Agent Index | GPT-5.6; Claude Fable 5 | Coding Agent Index; DeepSWE; Terminal-Bench | Follow-up/related AA posts on GPT-5.6 coding-agent evals | Low |
| 32 | ArtificialAnlys | @ArtificialAnlys | past month search result | Artificial Analysis AutomationBench-AA Zapier | Gemini; Claude; GPT | AutomationBench-AA | SaaS-workflow agent benchmark | Low |
| 33 | ArtificialAnlys | @ArtificialAnlys | past month search result | Artificial Analysis Harvey LAB-AA legal benchmark | Claude; GPT; GLM | Harvey LAB-AA | Legal-agent benchmark and model cost spread | Low |
| 34 | ArtificialAnlys | @ArtificialAnlys | past month search result | Artificial Analysis AA-Briefcase | Fable 5; GLM-5.2 | AA-Briefcase | Agentic business-work benchmark | Low |
| 35 | ArtificialAnlys | @ArtificialAnlys | past month search result | Artificial Analysis AA-Omniscience | Fable 5; GPT-5.5; Gemini | AA-Omniscience | Knowledge/hallucination benchmark | Low |
| 36 | ArtificialAnlys | @ArtificialAnlys | past month search result | Artificial Analysis CritPt physics benchmark | GLM-5.2; Opus 4.8 | CritPt | Scientific reasoning benchmark | Low |
| 37 | ArtificialAnlys | @ArtificialAnlys | past month search result | Artificial Analysis HLE benchmark model | GLM-5.2; GPT-5.5; Claude | HLE | Hard reasoning benchmark commentary | Low |
| 38 | lmarena_ai | @lmarena_ai | past month search result | LM Arena leaderboard GPT Claude Gemini | GPT; Claude; Gemini | Chatbot Arena; Arena leaderboard | Human preference/model-ranking updates | Low |
| 39 | lmsysorg | @lmsysorg | past month search result | Chatbot Arena leaderboard new model | frontier models | Chatbot Arena | Arena ranking update | Low |
| 40 | livebench_ai | @livebench_ai | past month search result | LiveBench leaderboard GPT Claude Gemini | GPT; Claude; Gemini | LiveBench | Contamination-resistant benchmark updates | Low |
| 41 | aider_chat | @aider_chat | past month search result | Aider Polyglot GPT Claude benchmark | GPT; Claude; Gemini | Aider Polyglot | Coding model benchmark updates | Low |
| 42 | paulgauthier | @paulgauthier | past month search result | Paul Gauthier aider benchmark Claude GPT | GPT; Claude | Aider Polyglot | Creator commentary on coding model performance | Low |
| 43 | SWEbench | @SWEbench | past month search result | SWE-bench GPT Claude benchmark | GPT; Claude; Gemini | SWE-bench Verified; SWE-bench Multilingual | Coding-agent benchmark updates | Low |
| 44 | terminal_bench | @terminal_bench | past month search result | Terminal-Bench GPT Claude benchmark | GPT; Claude | Terminal-Bench | Terminal-agent benchmark | Low |
| 45 | METR_Evals | @METR_Evals | past month search result | METR GPT-5.6 benchmark gaming | GPT-5.6 Sol | METR agent evals | Agentic benchmark gaming/behavioral caveats | Low |
| 46 | EpochAIResearch | @EpochAIResearch | past month search result | Epoch AI benchmark GPT Claude Gemini | frontier models | benchmark trends | Macro benchmark/progress analysis | Low |
| 47 | StanfordCRFM | @StanfordCRFM | past month search result | HELM Stanford leaderboard GPT Claude | GPT; Claude; Gemini | HELM; MedHELM | Holistic LLM eval framework updates | Low |
| 48 | huggingface | @huggingface | past month search result | Hugging Face Open LLM Leaderboard new model | open models | Open LLM Leaderboard | Open model leaderboard updates | Low |
| 49 | natolambert | @natolambert | past month search result | Nathan Lambert open model benchmark | Llama; Qwen; DeepSeek; GLM | open model benchmarks | Post-training/open-model evaluation commentary | Low |
| 50 | swyx | @swyx | past month search result | swyx GPT-5.6 benchmark eval | GPT-5.6; Claude | benchmarks/evals | AI engineering interpretation of benchmark results | Low |
| 51 | simonw | @simonw | past month search result | Simon Willison GPT-5.6 benchmark eval | GPT-5.6; Claude; Gemini | hands-on evals | Practical testing of model behavior | Low |
| 52 | HamelHusain | @HamelHusain | past month search result | Hamel Husain eval benchmark model | LLMs | evals | Product-eval methodology and model testing | Low |
| 53 | OpenRouterAI | @OpenRouterAI | past month search result | OpenRouter model benchmark price performance | frontier/open models | price/performance | Routing/API market signals | Low |
| 54 | paperswithcode | @paperswithcode | past month search result | Papers with Code leaderboard LLM benchmark | LLMs | leaderboards | Benchmark leaderboard amplification | Low |
| 55 | ScaleAILabs | @ScaleAILabs | past month search result | Scale Labs leaderboard model benchmarks | frontier models | Scale Labs leaderboards | Expert benchmark leaderboards | Low |
| 56 | ScaleAILabs | @ScaleAILabs | past month search result | Scale Labs DrugDiscoveryBench | GPT; Gemini; Claude | DrugDiscoveryBench | Life-science/drug discovery benchmark | Low |
| 57 | phylo_bio | @phylo_bio | past month search result | Phylo DrugDiscoveryBench benchmark | AI agents | DrugDiscoveryBench; BiomniBench | Biomedical agent benchmark | Low |
| 58 | OpenAI | @OpenAI | past month search result | OpenAI LifeSciBench GPT-Rosalind | GPT-Rosalind; GPT-5.5 | LifeSciBench | Life-science model benchmark | Low |
| 59 | OpenAI | @OpenAI | past month search result | OpenAI HealthBench GPT-5.6 | GPT-5.6 | HealthBench | Medical model benchmark | Low |
| 60 | OpenAI | @OpenAI | past month search result | OpenAI Terminal-Bench GPT-5.6 Sol | GPT-5.6 Sol | Terminal-Bench | Coding/terminal benchmark | Low |
| 61 | OpenAIDevs | @OpenAIDevs | past month search result | OpenAI Developers GPT-5.6 evals | GPT-5.6 | coding; tool use | Developer benchmark/performance details | Low |
| 62 | AnthropicAI | @AnthropicAI | past month search result | Anthropic Claude Sonnet 5 benchmark | Claude Sonnet 5 | coding; reasoning | Official Claude benchmark claims | Low |
| 63 | GoogleDeepMind | @GoogleDeepMind | past month search result | Google DeepMind Gemini benchmark | Gemini | reasoning; coding; multimodal | Official Gemini performance claims | Low |
| 64 | MistralAI | @MistralAI | past month search result | Mistral AI model benchmark | Mistral models | open/frontier benchmarks | Mistral model release performance | Low |
| 65 | AIatMeta | @AIatMeta | past month search result | Meta AI Llama benchmark | Llama | open model benchmarks | Llama benchmark/release comparisons | Low |
| 66 | deepseek_ai | @deepseek_ai | past month search result | DeepSeek model benchmark | DeepSeek | coding/reasoning benchmarks | DeepSeek model-performance posts | Low |
| 67 | Alibaba_Qwen | @Alibaba_Qwen | past month search result | Qwen benchmark model performance | Qwen | open model benchmarks | Qwen model benchmark results | Low |
| 68 | Zai_org | @Zai_org | past month search result | Z.ai GLM-5.2 benchmark | GLM-5.2 | AA Index; GDPval-AA | Source account for GLM-5.2 benchmark claims | Low |
| 69 | cursor_ai | @cursor_ai | past month search result | Cursor Grok 4.5 benchmark | Grok 4.5 | coding benchmarks | Coding-product benchmark context | Low |
| 70 | cognition_labs | @cognition_labs | past month search result | Cognition Devin benchmark | coding agents | SWE-bench | Coding agent performance | Low |
| 71 | windsurf_ai | @windsurf_ai | past month search result | Windsurf model benchmark coding | coding models | coding benchmark | AI IDE model-performance posts | Low |
| 72 | Replit | @Replit | past month search result | Replit Agent benchmark | coding agents | coding/task benchmarks | AI coding-agent performance | Low |
| 73 | LangChainAI | @LangChainAI | past month search result | LangChain agent benchmark | agent models | agent evals | Agent framework evaluation | Low |
| 74 | llama_index | @llama_index | past month search result | LlamaIndex RAG benchmark | LLMs | RAG evals | RAG/model evaluation | Low |
| 75 | modal_labs | @modal_labs | past month search result | Modal GPU inference benchmark | inference stacks | latency/throughput | Inference-performance benchmarks | Low |
| 76 | AnyscaleCompute | @AnyscaleCompute | past month search result | Anyscale LLM serving benchmark | LLMs | serving throughput | Infrastructure benchmarks | Low |
| 77 | togethercompute | @togethercompute | past month search result | Together AI benchmark throughput | open models | inference speed/cost | Model-serving benchmark posts | Low |
| 78 | FireworksAI_HQ | @FireworksAI_HQ | past month search result | Fireworks AI inference benchmark | open models | serving latency/throughput | Inference benchmark posts | Low |
| 79 | GroqInc | @GroqInc | past month search result | Groq inference benchmark | LLMs | tokens/sec latency | Inference speed benchmarks | Low |
| 80 | CerebrasSystems | @CerebrasSystems | past month search result | Cerebras inference benchmark | LLMs | tokens/sec | Inference speed benchmarks | Low |
| 81 | perplexity_ai | @perplexity_ai | past month search result | Perplexity model benchmark | Perplexity models | search/model quality | AI search model performance | Low |
| 82 | vellum_ai | @vellum_ai | past month search result | Vellum GPT-5.6 benchmarks explained | GPT-5.6 | benchmark analysis | Practical model comparison | Low |
| 83 | latentspacepod | @latentspacepod | past month search result | Latent Space AINews GPT-5.6 benchmarks | GPT-5.6 | benchmark roundup | Benchmark news synthesis | Low |
| 84 | deedydas | @deedydas | past month search result | Deedy Das model benchmark | GPT; Claude; Gemini | benchmark summaries | High-signal benchmark commentary | Low |
| 85 | AlphaSignalAI | @AlphaSignalAI | past month search result | AlphaSignal benchmark model | AI models | research benchmarks | AI R&D benchmark amplification | Low |
| 86 | cwolferesearch | @cwolferesearch | past month search result | Cameron Wolfe model benchmark | AI models | benchmark analysis | Technical paper/model benchmark breakdowns | Low |
| 87 | rasbt | @rasbt | past month search result | Sebastian Raschka model benchmark | open models | LLM benchmarks | Technical model comparison | Low |
| 88 | _philschmid | @_philschmid | past month search result | Philipp Schmid model benchmark | open models | leaderboards/inference | Hugging Face benchmark/inference testing | Low |
| 89 | Thom_Wolf | @Thom_Wolf | past month search result | Thomas Wolf open model benchmark | open models | leaderboards | Open-source model performance | Low |
| 90 | ClementDelangue | @ClementDelangue | past month search result | Clement Delangue open model benchmark | open models | leaderboards | Open model release/benchmark amplification | Low |
| 91 | abhi1thakur | @abhi1thakur | past month search result | Kaggle Abhishek model benchmark | ML models | Kaggle/evals | Applied model-performance tests | Low |
| 92 | valence_ai | @valence_ai | past month search result | Valence Labs Polaris benchmark | drug discovery models | Polaris | Drug-discovery benchmark platform | Low |
| 93 | wpwalters | @wpwalters | past month search result | Pat Walters ADMET benchmark | cheminformatics models | ADMET/blind challenge | Drug-discovery benchmark critique | Low |
| 94 | ProjectTDC | @ProjectTDC | past month search result | TDC benchmark model | therapeutic ML models | TDC | Therapeutics benchmark ecosystem | Low |
| 95 | openbioml | @openbioml | past month search result | OpenBioML benchmark | biology/chem models | BioML/Chem benchmarks | Open biology ML benchmark amplification | Low |
| 96 | chaidiscovery | @chaidiscovery | past month search result | Chai Discovery benchmark | structure models | protein/structure benchmarks | Biomolecular model performance | Low |
| 97 | RosettaCommons | @RosettaCommons | past month search result | Rosetta Commons benchmark | protein design models | protein modeling benchmarks | Protein modeling benchmark network | Low |
| 98 | iScienceLuvr | @iScienceLuvr | past month search result | MedARC Medmarks benchmark | medical LLMs | Medmarks | Medical LLM benchmark | Low |
| 99 | AISecurityInst | @AISecurityInst | past month search result | AI Security Institute Inspect Evals | frontier models | Inspect Evals | Model safety/eval framework | Low |
| 100 | ai_risks | @ai_risks | past month search result | CAIS WMDP benchmark | LLMs | WMDP | Bio/chem/cyber risk benchmark | Low |