AI300
  1. Zixuan Li / Lead Z.ai
    721.
    FrontierSWE v2: How far models can get within 20 hours https://t.co/usEFxjLXnj
    LLM score 25 · about 2 months ago
  2. Boris Cherny / Creator of Claude Code
    722.
    Fable 5.1 makes Claude Tag even more useful.›
    LLM score 72 · about 2 months ago
  3. Sholto Douglas / Researcher at Anthropic
    723.
    Excellent conversation - very direct, clear language.›
    LLM score 15 · about 2 months ago
  4. Shuchao Bi / Meta Researcher
    724.
    Significant jump on AA index. https://t.co/VzMLPRP3Vz
    LLM score 35 · about 2 months ago
  5. Thomas Wolf / Hugging Face Cofounder
    725.
    LLM training labs: the capabilities we are giving our models are god-like, they are now hacking into other companies and building hidden civilizations›
    LLM score 15 · about 2 months ago
  6. Myra Deng / Goodfire Researcher
    726.
    Can confirm @ren_hongyu https://t.co/zarTKB3Bfe
    LLM score 15 · about 2 months ago
  7. Jason Weston / Meta Research Scientist
    727.
    Muse Spark 1.3 is out, with strong performance.›
    LLM score 35 · about 2 months ago
  8. Jonathan Ross / TPU Creator
    728.
    Me vs. Service provider going back and forth for a week in email, summarized:›
    LLM score 15 · about 2 months ago
  9. Demis Hassabis / CEO of DeepMind
    729.
    Introducing Gemini 3.8 Flash, another upgrade in under a month! And the new 3.8 Flash Cyber pushes the frontier of cyber defense.›
    LLM score 25 · about 2 months ago
  10. Thang Luong / DeepMind Principal Scientist
    730.
    And we @GoogleDeepMind keep pushing the pareto frontier again with Gemini 3.8 Flash in less than 3 weeks, a major stepup from 3.7 in terms of agentic capabilities! More to come, stay tuned! https://t.co/S4Cqk3bpv2 https://t.co/wPr5munw5c
    LLM score 25 · about 2 months ago
  11. Andrew White / Edison Scientific Cofounder
    731.
    Look at those science evals! Great LABBench2 and biomystery bench scores! https://t.co/CYucFVqC9b
    LLM score 25 · about 2 months ago
  12. Zixuan Li / Lead Z.ai
    732.
    100M free GLM-5.3-Flash tokens on ZCode.›
    LLM score 15 · about 2 months ago
  13. Ben Burtenshaw / Hugging Face Researcher
    733.
    training agents, session 4 is live on september 10th, and it shows you how to build an RL environment for agents.›
    LLM score 72 · about 2 months ago
  14. Noam Brown / OpenAI Research Scientist
    734.
    Jakub is chief scientist at @OpenAI https://t.co/SAo1k01DFC
    LLM score 25 · about 2 months ago
  15. Jakub Pachocki / OpenAI Chief Scientist
    735.
    I want to prevent a race into unmonitorability kicked off by confused reporting.›
    LLM score 65 · about 2 months ago
  16. Shuchao Bi / Meta Researcher
    736.
    Looping and padding both increase the per token compute, which can enable the model to hide true intention in their explicit CoTs.›
    LLM score 65 · about 2 months ago
  17. Jerry Tworek / ex OpenAI VP of RL
    737.
    When I was young everyone repeated this meme that were only using 20% of our brain power.›
    LLM score 15 · about 2 months ago
  18. Mo Bavarian / OpenAI Research Scientist
    738.
    Sounds eerily like an agent chain of thought https://t.co/L8MR3dsgc3
    LLM score 15 · about 2 months ago
  19. Shuchao Bi / Meta Researcher
    739.
    try it out, pretty impressive https://t.co/1izhZMqblK
    LLM score 15 · about 2 months ago
  20. Merve Noyan / Hugging Face ML Engineer
    740.
    here's my take on consequences of chip bans and potential regulations, own your intelligence friends https://t.co/m2fWf9VJbI
    LLM score 35 · about 2 months ago
  21. Thomas Wolf / Hugging Face Cofounder
    741.
    RL explained with ducks https://t.co/uW0TP0fdpO
    LLM score 25 · about 2 months ago
  22. Shuchao Bi / Meta Researcher
    742.
    https://t.co/ja9lJ4ckxP https://t.co/JEUvTA8KuJ
    LLM score 15 · about 2 months ago
  23. Ilya Sutskever / Founder of SSI
    743.
    Neoclouds have limited cybersecurity.›
    LLM score 72 · about 2 months ago
  24. Yoshua Bengio
    744.
    In an interview in this article for The @guardian about AI deception, I explain why misaligned behaviors emerge from reinforcement learning, why they will continue to pose risks as models become more capable, and how we intend to rethink how we train AI systems at @LawZero_ .
    LLM score 82 · about 2 months ago
  25. Fei-Fei Li
    745.
    I'm so excited that our @theworldlabs team has achieved a major milestone today! Introducing Atlas - a first of its kind multimodal world model trained from scratch! 🚀›
    LLM score 35 · about 2 months ago
  26. Sharon Zhou / AMD VP of AI
    746.
    there is about to be a large wave of algorithms, architecture changes, and performance optimizations that trade more flops for less memory because the relative cost of flops per token to hbm per token is dropping.›
    LLM score 72 · about 2 months ago
  27. Ben Burtenshaw / Hugging Face Researcher
    747.
    shed >> office Most people don't know this, but over the last 3 years I've been working fully out of the shed at the end of my garden.›
    LLM score 15 · about 2 months ago
  28. Cameron Wolfe / Researcher at Netflix
    748.
    Ever wondered where the policy / importance ratio in the PPO loss function comes from?›
    LLM score 72 · about 2 months ago
  29. Zixuan Li / Lead Z.ai
    749.
    A Reset Card as a gift for GLM Coding Plan’s first birthday.›
    LLM score 15 · about 2 months ago
  30. Thomas Wolf / Hugging Face Cofounder
    750.
    seems like it https://t.co/EqAXmcP0zO https://t.co/K09zvpco1p
    LLM score 15 · about 2 months ago
More