What's new? Composer 1.5 uses 20x more RL steps and a thinking tokens system for code reasoning; it applies self ...

OpenAI is testing sponsored placements in ChatGPT for U.S. users on Free and Go tiers, with privacy rules, user controls, and clear ad labeling.

Meta AI is testing Avocado models, MCP integrations, and Manus browser agent support, with scheduled tasks and OpenClaw ...

Notion is testing new agent features, including a redesigned settings UI, new automation triggers, and upcoming Agents 2.0 ...

OpenAI introduces GPT-5.3-Codex with faster performance, expanded agent functions for developers, and strong coding benchmark ...

Google is testing a “Personal intelligence” layer in NotebookLM, hinting at personalized responses and custom prompts based on chat history.

What's new? Perplexity launches DRACO benchmark for AI research in law, medicine, finance and academia; it uses LLM as judge ...

What's new? Anthropic launches Claude Opus 4.6 for finance pros with beta Cowork for Mac, PowerPoint for Max, Team, Enterprise and Excel updates for paid subscribers; ...

Claude Sonnet 5 leaks suggest a new Anthropic model with strong math and coding capabilities, launching this week.

What's new? Max is a model router that auto-selects the best language model from over 5m community votes; the latency-aware ...

Anthropic appears to be preparing an upgraded voice mode for Claude across mobile and web. Earlier work in the mobile app ...

What's new? Step 3.5 Flash is a large language model using sparse experts and multi-token heads, deployable on NVIDIA and ...