ByteDance Seed and Tsinghua AIR introduced CUDA Agent, an agentic reinforcement learning system designed to generate high-performance GPU kernels, achieving significant speedups over torch.compile.
Nous Research updated Hermes Agent to v0.20.3, introducing 'Bot Mode' which allows users to organize agent profiles into a roster of named, interacting bots.
OpenAI Disbands AI Safety Team Amid Broader Restructuring SOFX
Nimbus builds production AI systems combining humans and AI end-to-end. From scoped pilot to production in 4 to 8 weeks.
Talk to Nimbus →We ran 904 DeepSWE rollouts on DeepSeek V4 Pro 0813 and GPT-5.6 Sol. Sol leads pass@1 by 10 points at 35x the cost; Pro wins pass@4, and a Pro-first cascade hits 83.0%.
No GPUs, no Agents, just really, really, really good infra and distribution.
DeepSeek AI has released DeepSeek Harness v0.1 in developer preview as an MIT-licensed, plugin-based agent harness that allows developers to customize models, tools, and loops without altering source code.
DeepSeek Releases V4-Pro Model, Hikes Prices Silicon UK
Simon Willison / Simon Willison's Weblog: Qwen 3.8 27B shows a 17GB open-weight general purpose model can have long context, effective tool calling, strong vision ability, and competent code generation — Friday's…
MiniMax released MiniMax-Music3, an open-weights text-to-music model capable of generating complete five-minute songs from lyrics and structured descriptions.
MiniMax Releases MiniMax-Music3: An Open-Weights Music Model Generating Complete Five-Minute Songs From Lyrics and a Structured Caption MarkTechPost
Apple Makes Major AI Strategy Shift in China, Develops Own LLM With Alibaba’s Support in Bid to Counter Huawei: Report Yahoo Finance
Claude to start watermarking AI-generated text – but will it make quality worse? The Guardian
Nvidia Invests $1.5 Billion In SB Energy And Backs 8-GW OpenAI AI Factory Campus In Ohio Pulse 2.0
Thomson Reuters: First Production-Ready Thomson LLM Delivers Very Strong Evaluation Results As Big 3 Growth Outlook Rises To 9.5%-10% Pulse 2.0
Bloomberg: Filing: Nvidia agrees to spend up to $105B to support SB Energy's new Ohio data center campus OpenAI is set to lease; Nvidia agrees to invest $1.5B in SB Energy — Nvidia Corp. has agreed to spend as…
Representing a significant milestone in AI-assisted mathematical research, a team at Axiom Math has automatically verified the proof of a theorem relating to prime numbers—colloquially referred to as the “246…
Alibaba answers Meta’s AI challenge with new laptop-ready model CNBC
Peking University and StepFun Unveil TensorCast: A Programmable Tensor Management Layer That Cuts LLM Time-to-First-Token by Up to 93.2% Pandaily
Anthropic's new invisible watermark marks content generated by AI chatbot Claude NPR
According to Bloomberg, Stripe is acquiring the AI startup OpenRouter for more than $7 billion, up from a latest valuation of $1.3 billion. OpenRouter offers access to over 400 AI models, has eight million users, and…
sycl: fix thread/block count in quantized cpy kernel launches (#27160) Adjusts the thread/block count to be proportional to the size of the quant, reducing under/over subscription. Largest perf improvement is the q4_0…
[SYCL] support OP OPT_STEP_ADAMW, OPT_STEP_SGD (#25268) fix conflict fix conflict of ops.md fix conflict of ops.md update the ops.md --------- Co-authored-by: Neo Zhang Jianyu Website: macOS/iOS: macOS Apple Silicon…
OpenAI expands the 1M token context window to ChatGPT subscribers for GPT-5.6 Sol KuCoin
Frontier AI models are actively breaching systems during testing, and enterprise security teams cannot ignore it MarketScale
Pro customers can now connect repositories to Vercel. Origin Once connected, pull requests on Origin will automatically generate Preview deployments on Vercel, and merging will trigger a Production deployment. You can…
We ran 904 DeepSWE rollouts on DeepSeek V4 Pro 0813 and Claude Fable 5. Fable leads pass@1 at 90x the cost; Pro wins pass@4, and a Pro-first cascade hits 82.7%.
Anthropic Lets Enterprises Bring Their Own Security to Claude — and That Changes How Companies Buy AI forkast.news
UPDATE: Anthropic Launches Claude Opus 5 For Lower-Cost Enterprise AI Shahbaz Ahmed (arc58nieFJ) Mshale
Google launches Gemini 3.7 Flash: What’s new in its latest AI model? The Indian Express
OpenAI's New Cyber Model Found Real Chrome Vulnerabilities — Now It Requires a Physical Security Key to Use Times Tabloid
Anthropic and OpenAI Revenues Soar, But a New Threat is Emerging Benzinga
Grok 4.6 matches GPT-5.6 Sol on benchmarks while undercutting its own $300 plan Startup Fortune
How OpenAI’s and Anthropic’s AI Models Went Rogue The Australian
OpenAI Previews GPT‑5.6 Sol Ultrafast Mode ManilaShaker Philippines
Anthropic's bio-weapons filter was down for nearly a year, exposing 133 million requests the-decoder.com
GPT-5.6 Vs Claude Fable 5: 7 Game-Building Tests (RAW RESULTS) Anthony Gordon (Q54sqs9yMQ) Mshale
How OpenAI's and Anthropic's AI Models Went Rogue MarketWatch
Anthropic develops Slack-like features for Claude Desktop TestingCatalog AI News
Anthropic AI Confusion at Pentagon: Nuclear-Level Threat dailybeirut.com
Anthropic explains how Claude's AI text watermark will work Storyboard18
In a safety report, Anthropic reveals that its internal filtering system for biological and chemical weapons risks was inactive for nearly a year. During that time, around 50,000 external feedback contractors ran about…
Anthropic Finds AI Agents Disabling Rivals, Evading Safety Restrictions Benzinga
Reuters: Z.ai says GLM-5.3 scores 84.5% on CyberGym, vs. Mythos 5's 83.8%, and its most sensitive cybersecurity functions will only be available to verified users — Chinese AI startup Z.ai said on Friday its…
OpenAI launches GPT-5.6 with a 25x cost reduction for the smallest model. KuCoin
10 most important shipping events. Three minutes to read. One-click unsub.