OpenAI is expected to shift its revenue structure from the high-priced Plus subscription to a lower-cost, ad-based model.
Chrome has released an on-device Prompt API based on Gemini Nano as an origin trial.
Mintlify has launched Editor, featuring WYSIWYG, real-time collaboration, and Git sync.
An Abstract CoT technique has been proposed that allows LLMs to reason efficiently using meaningless tokens.
SOB unveiled a new benchmark that verifies not just JSON schema pass rate but also value accuracy.
As conflict between the US and China over AI technology theft and investment regulations intensifies, China's DeepSeek is narrowing the gap with low-cost, high-efficiency models.
Warp has open-sourced its client and announced an 'agent-first' development model in which agents handle implementation.
OpenAI Codex's GPT-5.5 instructions include a rule to avoid mentioning unrelated animals or creatures.
An open-source macOS app has been released that shows Claude Code's 5-hour and weekly usage in real time from the menu bar.
Storybloq, a project dashboard for Claude Code, has been listed on the Mac App Store.
Nvidia has unveiled Nemotron 3 Nano Omni, an open-weight multimodal model.
llama.cpp has added DKQ=320/DV=256 flash-attn CUDA support for Mistral Small 4.
Amazon revealed that it will co-develop an AWS-based AI agent platform with OpenAI.
OpenAI has unveiled Symphony, an engineering agent framework that autonomously carries out project work.
Anthropic has unveiled Claude connectors for creative tools such as Blender and Adobe.
Dreambase has launched a 'Skills' feature that helps AI agents understand data and business logic.
SOB has released a new benchmark that measures not only JSON parsing but also value accuracy.
On an M2 MacBook Pro with 32GB RAM, Qwen 3.6 27B's processing speed plummeted to 4t/s at a 52,000-token context.
Mistral has released Workflows, an enterprise AI orchestration layer, in public preview.
In 8x H100 experiments, DeepSeek V4-Flash's mHC was found to become redundant after 3 layers and concentrate magnitude solely on BOS.
The pipeline structures of 6 Codex CLI workflows were compared in a single table.
An incident occurred in which an operational database was entirely deleted while using the AI coding tool Cursor.
A vLLM PR has added support code for Mistral-Medium 3.5 128B.
Vercel has unveiled Native Deployment Checks, available to all teams, which run lint and typecheck in parallel with every deployment.