Model Follower.

Up to date

Rebuilt
2026-10-25 06:54 UTC
Release radar
117 on watch
Hub snapshot
2026-10-25
Re-checked
daily · every number sourced and dated

Links / Past days

Past days

Each day's most-shared AI links, as the board stood at the end of that day (UTC). A day is sealed once it ends and is kept as it was.

Days 44 shown Rebuilt 37 early days from issues on file
  1. Sun 25 Octtoday
    1. No Signs of AI Payoff Outside the Tech Sector apollo.com · 2
    2. Zagitta/gpu-unlock: Unlock FP/BF16 instruction issue rate on Nvidia 3000's cards to their professional equivalents github.com · 3
    3. Comfy-Org/workflow_templates: ComfyUI template workflows github.com · 1
  2. Sat 24 Oct
    1. Kimi K3.1 Pro Tech Blog: Advancing Open Intelligence kimi.ai · 5
    2. No Signs of AI Payoff Outside the Tech Sector apollo.com · 2
    3. Zagitta/gpu-unlock: Unlock FP/BF16 instruction issue rate on Nvidia 3000's cards to their professional equivalents github.com · 3
  3. Fri 23 Oct
    1. Kimi K3.1 Pro Tech Blog: Advancing Open Intelligence kimi.ai · 4
    2. Introducing Claude Duet claude.com · 5
    3. LLMs are making formal software verification feasible at a much larger scale anthropic.com · 5
  4. Thu 22 Oct
    1. OpenAI models are coming to Google Cloud openai.com · 5
    2. OpenAI’s model helps catch critical PHP flaws openai.com · 4
    3. Introducing Claude Duet claude.com · 3
  5. Wed 21 Oct
    1. OpenAI Acquires Orchestra Bio to Accelerate Drug Development openai.com · 4
    2. How Claude designed protein binders against scorpion neurotoxins anthropic.com · 4
    3. Meta's Muse hits 10M US downloads in 40 days, a milestone reached by only 11 other apps techcrunch.com · 3
  6. Tue 20 Oct
    1. of Homeland Security formally banned DeepSeek from civilian agency systems, giving departments 60 days to request… public-inspection.federalregister.gov · 4
    2. Diffusion at Scale: Introducing Mercury 3 inceptionlabs.ai · 4
    3. MiniMax M3.1 Flash: Redefining Agent Performance and Cost minimax.io · 3
  7. Mon 19 Oct
    1. Investing in the public benefit anthropic.com · 5
    2. DeepSeek-V4.1-Flash-1017 (Hugging Face Repo) huggingface.co · 4
    3. How the Claude Code team does code review / claude.dev Blog claude.dev · 4
  8. Sun 18 Oct
    1. Introducing Muse Spark Open, The Most Capable US Open‑Weight Model to Date research.meta.ai · 7
    2. Release: llm-tools-datasette 0.1.1 When I switched LLM over to httpx2 in LLM 0.33 I forget that this would break any… github.com · 1
    3. When I switched LLM over to httpx2 in LLM 0.33 I forget that this would break any plugins which used httpx without… github.com · 1
  9. Sat 17 Oct
    1. Introducing Muse Spark Open, The Most Capable US Open‑Weight Model to Date research.meta.ai · 6
    2. Inside Microsoft's Frontier Pre-Training Stack microsoft.ai · 3
    3. Introducing Mistral Large 4 mistral.ai · 12
  10. Fri 16 Oct
    1. Introducing Muse Spark Open, The Most Capable US Open‑Weight Model to Date research.meta.ai · 5
    2. Anthropic update on independent evaluators anthropic.com · 3
    3. How the Claude Code team does code review / claude.dev Blog claude.dev · 2
  11. Thu 15 Oct
    1. Redesigning source control at Anthropic anthropic.com · 9
    2. Introducing Grok Voice 3.0 x.ai · 2
    3. Jev: For the builders AND dreamers typesafe.ai · 2
  12. Wed 14 Oct
    1. Gemini 4 Argon: Built to solve your hardest problems blog.google · 5
    2. LFM2.5-24B-A2B: Built for Fast On-Device Agents liquid.ai · 4
    3. Strengthening our defenses against distillation attacks anthropic.com · 5
  13. Tue 13 Oct
    1. Muse Wonder: A Frontier Model For Personal & Professional AI Agents research.meta.ai · 7
    2. Redesigning source control at Anthropic anthropic.com · 3
    3. AI models may be capable of self-replicating without access to a datacenter anthropic.com · 4
  14. Mon 12 Oct
    1. Microsoft-Decision-1 commandline.microsoft.com · 5
    2. Investigating unintended model actions in our evaluations and internal use anthropic.com · 5
    3. Muse Wonder: A Frontier Model For Personal & Professional AI Agents research.meta.ai · 2
  15. Sun 11 Oct
    1. EmbeddingGemma 2: an open, lightweight multimodal embedding model blog.google · 6
    2. TypeSafe A raises Series AI typesafe.ai · 4
    3. Long-WAM: Scaling the Context of World-Action Models arxiv.org · 2
  16. Sat 10 Oct
    1. Python 3.15.0 added to actions/python-versions github.com · 1
    2. Tags chatgpt.com · 1
    3. Now that this has landed, you can add "3.15" to a GitHub Actions testing matrix to run tests against the new Python… python.org · 1
  17. Fri 9 Oct
    1. Introducing Claude Haiku 5.5 anthropic.com · 11
    2. Introducing the Anthropic Cyber Mission anthropic.com · 4
    3. GPT-6 and Intelligent UI for everyone openai.com · 7
  18. Thu 8 Oct
    1. Introducing Claude Haiku 5.5 anthropic.com · 10
    2. GPT-6 and Intelligent UI for everyone openai.com · 7
    3. Playground labs.google · 3
  19. Wed 7 Oct
    1. Sharing AI progress in mathematics openai.com · 9
    2. Introducing Mistral Large 4 mistral.ai · 6
    3. EmbeddingGemma 2: an open, lightweight multimodal embedding model blog.google · 5
  20. Tue 6 Oct
    1. Introducing Beam: Reflection’s 501B open-weight model reflection.ai · 7
    2. OpenAI to Add Text Watermarks openai.com · 3
    3. EmbeddingGemma 2: an open, lightweight multimodal embedding model blog.google · 2
  21. Mon 5 Oct
    1. Solving Open Research Problems Together research.meta.ai · 3
    2. Introducing Beam: Reflection’s 501B open-weight model reflection.ai · 2
    3. Anthropic seeks religious wisdom for raising Claude nytimes.com · 5
  22. Sun 4 Oct
    1. Qwen3.8 27B addition in words github.com · 1
    2. Colin Frasier posted on Bluesky about an experiment he ran over two years ago using GPT-4o to see how well it could… bsky.app · 1
    3. Here's a version of the report that includes the reasoning traces from some of those larger calculations, which include… gist.github.com · 1
  23. Sat 3 Oct
    1. Google Cloud launched a similar feature in July, called Spend Caps, which lets you "set a monthly financial cap on… cloud.google.com · 1
    2. I just sent the September edition of my sponsors-only monthly newsletter. github.com · 1
    3. If you are a sponsor (or start a sponsorship now) you can access it here. github.com · 1
  24. Fri 2 Oct
    1. Gemini 4 Argon: our next era of frontier intelligence blog.google · 9
    2. Our first streaming transcription model debuts at no. 1 on Artificial Analysis microsoft.ai · 3
    3. Tavus’ Griffin passes for human on video calls tavus.io · 3
  25. Thu 1 Oct
    1. Gemini 4 Argon: our next era of frontier intelligence blog.google · 7
    2. Dots. OpenAI has given us… dots openai.com · 10
    3. GLM-5.3 and the spread of advanced cyber capabilities anthropic.com · 8
  26. Wed 30 Sept
    1. GPT 6.1 Sol: Near-Astra intelligence for a fifth of the price openai.com · 9
    2. OpenAI introduces "Dots," AI-driven agents powered by GPT-6 Astra, designed to handle tasks autonomously using their… openai.com · 8
    3. GLM-5.3 and the spread of advanced cyber capabilities anthropic.com · 6
  27. Tue 29 Sept
    1. Introducing Claude Sonnet 5.5 anthropic.com · 7
    2. Launching Meta Enterprise Platform about.fb.com · 4
    3. NVIDIA Open Agent Safety Platform: A Reference for Continuous In-Silicon Agent Monitoring developer.nvidia.com · 4
  28. Mon 28 Sept
    1. Read the full article here blogs.microsoft.com · 4
    2. Then Madison Mills reported in Axios that we can raise the stakes axios.com · 4
    3. Anthropic Signed an $11.6 Billion Akamai Compute Deal ir.akamai.com · 3
  29. Sun 27 Sept
    1. Introducing Claude Opus 5.5 anthropic.com · 10
    2. said Devin crossed a $1B annualized revenue run rate less than two years after general availability. cognition.com · 2
    3. DeepSeek Elastic Compute (DSec): A Sandbox Infrastructure for Effective Agentic Training at Scale arxiv.org · 2
  30. Sat 26 Sept
    1. OpenAI says Astra was not involved. openai.com · 4
    2. In late July, a UK government cybersecurity lab caught an AI agent inventing fake identities to trick a real person… aisi.gov.uk · 3
    3. AI agents being tested by OpenAI involved in cyber-attack on another service, say researchers theguardian.com · 2
  31. Fri 25 Sept
    1. Behind Project Suncatcher, our moonshot to put AI in space blog.google · 5
    2. Introducing Claude Opus 5.5 anthropic.com · 9
    3. Claude discovers a novel enzyme system anthropic.com · 8
  32. Thu 24 Sept
    1. Claude discovers a novel enzyme system anthropic.com · 7
    2. GPT-6 Sol and Luna openai.com · 10
    3. OpenAI gives us MentalHealthBench openai.com · 4
  33. Wed 23 Sept
    1. GPT-6 Sol and Luna openai.com · 8
    2. Introducing Claude Opus 5.5 anthropic.com · 7
    3. Introducing Grok 4.7 x.ai · 8
  34. Tue 22 Sept
    1. Introducing Grok 4.7 x.ai · 6
    2. OpenAI answered with GPT-6 Sol and Luna openai.com · 2
    3. Introducing Claude Opus 5.5 anthropic.com · 2
  35. Mon 21 Sept
    1. Astra for Law openai.com · 7
    2. Alibaba released Qwen-Image-2.1, an open weights image generation and editing model with 7 billion parameters in its… qwen.ai · 3
    3. Manus Targets $4B Valuation techcrunch.com · 2
  36. Sun 20 Sept
    1. Qwen Image 2.1 qwen.ai · 2
    2. Now, Figure AI just dropped Helix 2.5, a new model that enabled its humanoid robots to walk into 30 homes they’d never… figure.ai · 2
    3. Welcome back, Superhuman. humansecurity.com · 1
  37. Sat 19 Sept
    1. Measurements for understanding the pace of AI development inside frontier labs anthropic.com · 3
    2. Noam Brown – Agent swarms, alignment, & recursive self-improvement dwarkesh.com · 2
    3. My default expectation (absent an extensive pause) is that a similar thing will happen: they’ll turn out to be lesswrong.com · 1
  38. Fri 18 Sept
    1. OpenAI’s new rules for reporting model misbehavior openai.com · 4
    2. Introducing System One Models & Jev typesafe.ai · 5
    3. Toward Recursive Self-Improvement: How GLM Built Its Own Inference Infrastructure z.ai · 3
  39. Thu 17 Sept
    1. Claude Cowork and chat are now one Claude claude.com · 5
    2. Your AI agents can now control your Google Home devices techcrunch.com · 2
    3. Introducing System One Models & Jev typesafe.ai · 4
  40. Wed 16 Sept
    1. Introducing System One Models & Jev typesafe.ai · 3
    2. Introducing Odyssey-3: A General-Purpose Physical Intelligence odyssey.systems · 2
    3. Meta One is a subscription service that offers enhanced AI and self-expression features across Instagram, Facebook, and… about.fb.com · 2
  41. Tue 15 Sept
    1. Dario Amodei darioamodei.com · 8
    2. David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models x.com · 4
    3. the model behind ChatGPT’s new Voice mode. openai.com · 3
  42. Mon 14 Sept
    1. Dario Amodei darioamodei.com · 7
    2. and NASA released the NASA-IBM Lunar Foundation Model, an open-source model combining decades of lunar data from US and… research.ibm.com · 2
    3. ToolGrad: Efficient tool-use dataset generation with textual “gradients” research.google · 2
  43. Sun 13 Sept
    1. Dario Amodei darioamodei.com · 2
    2. Regent is pitching it as a new mode of coastal transport, with proposed routes already planned between cities like New… smartcitiesdive.com · 1
    3. Garry Tan wants US open-weight AI labs to 'distill' frontier models, too techcrunch.com · 1
  44. Sat 12 Sept
    1. Countering misuse of AI: September 2026 / Anthropic anthropic.com · 5
    2. Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases withspecific.com · 1
    3. This seems more probable than it was before. reddit.com · 1