Daily AI news, sourcedUpdates / Models
Updates Models · 505 items · newest issue first · page 23 of 26
AreaAllModelsAgents & codingApps & productsChips & infraResearchPricing & limitsBusiness & fundingPolicy & safetyOther
- Gliner2.5-Decide: a Jev-style decision model in community testing
- Gemini 4 Argon appears in community testing
- Google surpassing OpenAI in VAls RSI Index, community claims
- Introducing Gemini 4 Argon
- M5 Ultra local test: Qwen3.8 Flash Next vs Laguna S 2.1
- ATX-Swift-1.5-Qwen3.8-27B-Uncensored community quantization sustaining 50-65 t/s
- GPT-6.1 Sol reaches third place in Humanity's Last Exam — Diamond
- Anthropic lists introducing docs for Claude Fable 5 and Mythos 5
- Anthropic adds thinking-output docs for Claude Fable 5 and Mythos 5
- Cohere releases Embed v5 embedding model
- OpenAI releases GPT-6.1 Sol at one-fifth of Astra's price
- Anthropic's Mike Krieger: Haiku 5.5 will round out the family in the coming weeks
- NVIDIA highlights Korean team's top-5 open model score on Artificial Analysis
- Artificial Analysis index ranks GPT-6.1 Sol behind Opus 5.5, Sonnet 5.5 and Fable 5.1
- Test code in DeepSeek's harness repo hints at DeepSeek-V4-Mini with 1M context
- Artificial Analysis launches new Pareto line featuring GPT-6.1 Sol
- Anthropic site lists Claude Fable and Mythos 5.1 pages
- Community model claims better calibration than Jev with multi-modal abilityUnconfirmedreddit-machinelearningSep 30Models
- 500M VLM answers typed questions about images with calibrated confidenceUnconfirmedreddit-machinelearningSep 30Models
- Community tests 5 Qwen3.6-35B-A3B finetunes against the base modelUnconfirmedreddit-localllamaSep 30Models