3+ day, 21+ hour ago   (255+ words) Mercury 2.5 is now available through Inception, Baseten, and OpenRouter, with Mercury Voice and Mercury Router entering preview. Mercury 2.5 generates 1,107 tokens per second on widely available NVIDIA GPUs and supports a 260K-token context window. Its capabilities include tunable reasoning, parallel tool…...

5+ day, 21+ hour ago   (157+ words) Managed Agents on AOI Platform demo Managed agents will not be the only announcement at DevDay. Model updates, security solutions, and more are also expected. Although access to the agents UI is not available yet, the codebase suggests functionality very…...

6+ day, 14+ hour ago   (386+ words) TestingCatalog AI News The Institute of Foundation Models has released K2 Horizon, a connected family of six language models ranging from 0.9 billion to 375 billion parameters, published alongside details on how each was trained. The sizes run 0.9B, 3.7B, 7B, 32B, 36B, and 375B, and IFM maps each…...

1+ week, 1+ day ago   (170+ words) GPT-6 Astra is rolling out to ChatGPT paid plans and major cloud APIs with faster computer use, new Codex context memory and stronger safeguards. GPT-6 Astra is now available to all Pro, Enterprise, and Business Premium users in ChatGPT Work…...

1+ week, 3+ day ago   (139+ words) OpenAI says Astra is its first model to reach the Critical cybersecurity threshold, with limited alpha access planned before wider Daybreak Blue use. In hands-on assessments, Astra built a browser compromise chain that escaped the sandbox and ran host commands…...

1+ week, 4+ day ago   (155+ words) Fable 5.1 is now broadly available with 75% cheaper cache reads, while Mythos 5.1 is limited to vetted cyber and life-science users. Cache reads now cost 75% less at $0.25 per million tokens. Anthropic estimates typical costs will fall 25%, with savings on highly agentic work…...

2+ week, 19+ hour ago   (276+ words) The latest surfaced outputs were reportedly generated zero-shot with Max effort, where Astra spends substantially longer reasoning than GPT-5.6 Sol. OpenAI appears to be moving Astra closer to release, with reports pointing to expanded internal testing and a new checkpoint…...

2+ week, 5+ day ago   (254+ words) The latest Gemini desktop app update introduces Avatars and hints at future app and plugin discovery features, as Google readies Gemini 4. At the same time, Google is preparing internally for Gemini 4. Based on earlier information about roughly six-month training cycles,…...

3+ week, 4+ day ago   (252+ words) Anthropic is testing Hub Mode in its mobile app, hinting at a future tool for managing agents, though its release timeline is still unknown. Anthropic is working on a new feature called Hub Mode, recently spotted in its mobile app....

3+ week, 6+ day ago   (292+ words) Anthropic has updated its upcoming collaborative projects for Claude Code. During project creation, users can now add repositories directly as persistent context. The interface recommends adding only the repositories required for each session, and suggests attaching additional repos on a…...