AI / News
What the labs announced
Official blogs and news pages of more than a dozen AI companies, checked every 6 hours. Titles are as published; links go to the official pages.
Updated 11-21 18:50 UTC
Saturday 21 November
Friday 20 November
Google GeminiGet your year-end tasks done with new Connected Apps.
OpenAICombatting adversarial distillation
AnthropicAnthropic announces closing of initial public offering, including full exercise of underwriters’ option to purchase additional shares, and closing of concurrent
Hugging FaceIntroducing harness adapters: adapt your favorite model to your harness to improve performance and costRelease
OpenAIHebbia sees improvements in financial research with GPT-6 Astra
OpenAICan AI help scientists identify promising drug candidates?
Hugging FaceWelcome Inkling 2 Mini, Cutting-edge Performance for your Local Setup
Thursday 19 November
Google GeminiICYMI: Transfer your history from other AI apps into Gemini.
Google 开发者Multi-Turn Agentic Reinforcement Learning with Tunix and Gemma 4 on Google TPUs
Google AIIntroducing Tool Retrieval in the Gemini APIRelease
Google GeminiLet Gemini tackle repetitive chores on your Mac
Google AIWalter Isaacson on curiosity and creativity in the AI age.
OpenAIHow OpenAI’s Performance Factory keeps ChatGPT fast
OpenAIDeploying Daybreak Red
OpenAISee PR scans in action
OpenAIExplore the Codex Security Plugin
OpenAIGet started with Daybreak Blue
Wednesday 18 November
GitHub CopilotConnect Supabase, Box, Calendly, and more to Copilot
GitHub CopilotGemini 4 Neon in GitHub Copilot
Google GeminiIntroducing Gemini 4 NeonRelease
Hacker NewsGoogle Gemini's "Tasks" Personal Assistant Is Now FreeCommunityDeal
Hugging FaceDaisyRose-27B: Pushing the Frontier of Post-Training Efficiency
Hugging FaceWhy your fine-tuning job takes longer than the FLOPs say
Google GeminiGenie 4 is now availableRelease
Google GeminiThe Gemini app is now available on lower-memory Android devices.Release
GitHub CopilotCustomize GitHub Copilot for your team
Hugging FaceTime-Series Models in the Streamhouse: A New Frontier and Use Cases
OpenAIInside the Codex sandbox
OpenAIResearch acceleration: The jagged frontier
OpenAIHow Sumitomo Life’s leaders helped teams adopt ChatGPT Enterprise
OpenAIManaging frontier AI risk: From training to deployment
Google 开发者Build full-stack Android apps with Gemini in Google AI Studio
AnthropicAnthropic prices initial public offeringPricing
OpenAIThe Silent Consensus
OpenAIAdvancing chip design with OpenAI models
OpenAILe Groupe Figaro and OpenAI partner to bring Le Figaro journalism to ChatGPT
Hugging Face180B Parameters, 32GB of RAM: Pushing MoE Inference to the Limit with Transformers
Hugging FaceRumble-2-Flash: an open 0.5B speech model for 40+ languages
Tuesday 17 November
Google GeminiSee what 5 builders are making with Gemini 4 Carbon
GitHub CopilotOpenAI’s GPT-6.2 Astra and GPT-6.2 Sol now availableRelease
Google 开发者Gemini 4 Hydrogen: a faster model for general purpose robotics
Together AILast year we built megakernels. This year we deleted them: The road to Together Compiler
Together AIMaking open models work for long-horizon agentic workloads
GitHub CopilotGitHub Copilot Fusion updates: more availability, benchmarks, and models
GitHub CopilotGitHub Copilot app, Microsoft Ignite releasesRelease
GitHub CopilotCopilot code review: Adaptive effort, secret detection, and more
GitHub CopilotMAI Code 2 and MAI Code 2 Flash in GitHub Copilot
GitHub CopilotGitHub Copilot in JetBrains, Ignite releasesRelease
GitHub CopilotMonitor GitHub Copilot activity through Microsoft Agent 365
GitHub CopilotGitHub Copilot CLI adds Microsoft Entra sign-in as a previewRelease
GitHub CopilotTrack Copilot code review outcomes in Repository insights
GitHub CopilotGitHub Copilot Evals is now in public previewRelease
OpenAIIntroducing GPT-6.2 Astra and SolRelease
Google GeminiPut Gemini to work for you
OpenAIIntroducing ChatGPT for SalesRelease
Google 开发者TPU Model Design Considerations and Optimization Guide
OpenAIStrengthening NATO and allied cyber defenses with AI
OpenAIIntroducing OpenAI for ChileRelease
OpenAIAstra for Law leads in legal intelligence
Monday 16 November
GitHub CopilotGitHub Copilot weekly releases — November 9Release
GitHub CopilotGitHub Copilot models deprecated on November 16, 2026
GitHub CopilotControl where GitHub Copilot customizations come from
GitHub CopilotVS Code retires the Local harness in February 2027
Google 开发者Agents are sessions, not requests: size GKE node pools for AI agent sessions
Google 开发者Securing Your Agents: A Hands-On Guide to Agent Anomaly Detection
Google GeminiRegional voices are rolling out in the Gemini app.
OpenAIBroadening access to AI for scientists
OpenAIStrengthening the fight against online child sexual abuse with new tools and support
GitHub CopilotTake GitHub Mobile with you on iPhone Duo
Google AINew Ask Maps features to help with holiday travel and prep
OpenAIScience at the Frontier: How scientists use ChatGPT
OpenAIHow Revolut raises ambition without lowering the bar
OpenAILVMH’s Media Division announces strategic partnership with OpenAI
OpenAIOur first phase of embedded assessments
OpenAIAdvancing computer-use agents for healthcare teams
OpenAIMore ways to move care forward with ChatGPT for Healthcare
OpenAIThe King’s Trust and OpenAI launch £1m partnershipRelease
OpenRouterOpenRouter Now Powers Inference on Hugging Face
Hugging FaceHugging Face and OpenRouter join forces to make open-weight models easy to runRelease
OpenAIAt Holiday Extras, dots get to work before anyone asks
OpenRouterTighter Access Controls for Management Keys
Sunday 15 November
Saturday 14 November
- Hacker NewsShow HN: Free Dictionary API – Wiktionary data, no API key requiredCommunityDeal
Friday 13 November
GitHub CopilotUpcoming deprecation of selected GitHub Copilot models in December
DeepSeekThe full DeepSeek App experience, now via API
GitHub CopilotGitHub Enterprise Server support in the GitHub Copilot app
GitHub CopilotGemini 4 Hydrogen in GitHub Copilot
Hugging FaceBuild a Robot Pick-and-Place Simulation with SimReady Assets
Hugging FaceXYBench: Helping with the problem behind the question
Google GeminiConnect Google Wallet to Gemini for personal finance insights
OpenAIIntroducing ChatGPT WalletRelease
Hugging Facellama.cpp v1.0.0 : A stronger foundation for the local AI ecosystem
OpenAIShield helps technicians hand off legacy portal work with OpenAI
Google 开发者Catch Go bugs that fuzzing misses with Gemini and a constraint solver
Google 开发者How Agent Anomaly Detection scores on benchmarks we didn't write
OpenRouterReranking in RAG: How to Fix Bad Retrieval with a Reranker
Thursday 12 November
OpenAIHow Amplitude ships product improvements in 1–2 days with OpenAI
Hugging FaceLLMON: An LLM-Native Markup Language, now in Granite 4.1 3B and 8B
Google GeminiGoogle Search gets a major upgrade with Gemini 4 Hydrogen
Google GeminiGemini 4 Hydrogen: everyday coding and agentic workflows for developers
Google GeminiIntroducing Gemini 4 HydrogenRelease
MicrosoftFrom AI ambition to business impact
Google GeminiGet quick answers from Gemini on any website.
MicrosoftScaling AI value without losing control: Lessons from Frontier Firms
OpenAIGondola doubles revenue as its team builds more with Codex
OpenAIHoliday Extras is rethinking what it can build with Codex
OpenAIGilbert + Tobin rethinks the business of law with AI
OpenRouterHow to Stop an AI Agent Stuck in an Infinite Loop
OpenRouterHow to Benchmark LLMs
OpenRouterSelf-Hosted LLM vs API: When Does Self-Hosting Pay Off
OpenRouterHow We Built a Self-Correcting Localization System for OpenRouter
OpenAISupporting research on AI and children’s safety
Hugging FaceThe Definitive Fine-Tuning Guide for 3D Generation Models
OpenAIChatGPT Ads increases its market presence in 44 countries
Wednesday 11 November
Hacker NewsGemini Decision 1 is Free on Google AI StudioCommunityDeal
Google AIGemini Decisions API: Fast, calibrated multimodal decisions
Google GeminiTalk it through with Gemini, now on Apple CarPlay
OpenRouterLLM Quantization: What fp8 and fp4 Do to Quality and PricePricing
Tuesday 22 September
OpenRouterBatch API: half-price inference by bundling requestsPricing
Earlier (already on the sites when we started)
DeepSeekDeepSeek API introduces Context Caching on Disk, cutting prices by an order of magnitudePricingRelease
Nothing in this category lately.