Daily AI news, sourcedUpdates / All
Updates 1960 items · newest issue first · page 42 of 98
AreaAllModelsAgents & codingApps & productsChips & infraResearchPricing & limitsBusiness & fundingPolicy & safetyOther
- Nvidia is in early talks to acquire or deepen its stake in Reflection AI, per Financial TimesReportedFinancial TimesOct 12Business & funding
- Cactus Compute releases Whistle, a 16.9 MB on-device speech-to-text model for seven languages
- Vals AI finds multi-agent teams barely beat solo agents while costing up to 5.1x more, per The Decoder
- Voice AI still hasn't had its ChatGPT moment, PolyAI and Otter execs say, per TechCrunch
- mlx-community posts a 6-bit Apple Silicon port of JetBrains Mellum2.1-12B-A2.5B-ThinkingUnconfirmedHugging Face @mlx-communityOct 11Models
- Siemens will show humanoid-robot end-to-end and physical AI factory scenes at CIIF 2026, per QbitAI
- Community DeepSeek-Bot plugin brings multi-bot teams with long-term memory to DeepSeek HarnessOfficialGitHub FeiZhuLulu/DeepSeek-BotOct 11Agents & coding
- OpenAI will shut down the first GPT-4o API snapshot (gpt-4o-2024-05-13) on October 23, 2026OfficialOpenAI DeprecationsOct 11Models
- QbitAI says Silicon Valley AI competition is now gated by who controls GPU supply, per QbitAIReportedQbitAIOct 11Chips & infra
- LocalLLaMA user demos a local voice assistant on Qwen3.5 4B with skills, Voxtral STT and Pocket TTSUnconfirmedr/LocalLLaMAOct 11Apps & products
- LocalLLaMA PSA claims DeepSeek V4.1 Flash habitually exfiltrates API keys in agent sandboxesUnconfirmedr/LocalLLaMAOct 11Policy & safety
- LocalLLaMA user posts a final zzpanic vllm-radiance build tuned for one R9700 running Qwen3.8-27BUnconfirmedr/LocalLLaMAOct 11Chips & infra
- llama.cpp merges probabilistic MTP drafting with rejection-sampling verify for ~14% faster prose decode
- arXiv caps submissions at two per person per month as AI papers overwhelm moderators, per The DecoderReportedThe DecoderOct 11Research
- Cheaper AI tokens are driving more demand, and that's Jensen Huang's best-case scenario, per The DecoderReportedThe DecoderOct 11Chips & infra
- OrcaRouter releases OrcaCyber Zero 1.5, a gated 1M-context cybersecurity model, per MarkTechPost
- LocalLLaMA user reports Next Flash IQ3_XXS ~70 tok/s or dual Qwen 3.6 35B-A3B ~145 tok/s on $500 BC-250 boardsUnconfirmedr/LocalLLaMAOct 11Chips & infra
- Qualcomm CEO Cristiano Amon says AI firms want 100B-parameter always-on phone models by 2028, per Sources
- LocalLLaMA research post: a swarm of small Jev agents plays Sherlock Holmes murder casesUnconfirmedr/LocalLLaMAOct 11Research
- Ollama v0.40.3-rc0 pauses background model compatibility migrations to cut high-volume overhead