Recursive’s $410M AWS Deal: Self-Improving AI Explained
Recursive Superintelligence signed a $410M AWS deal to scale self-improving AI. Here is what the benchmarks show and what developers should expect by ...
Claude Opus 5 Is Out: Migrate from Opus 4.8 Now
Claude Opus 5 launched July 24 with two API breaking changes. Here's what breaks, benchmark ...
DeepSeek V4 Pro: 80.6% SWE-Bench at $0.87/M Output
DeepSeek V4 Pro ties Gemini 3.1 Pro on SWE-bench Verified at $0.87/M output — 28.7x ...
AI Agents Skip Policy Rules 78% of Tasks—New Data
HANDBOOK.md shows no frontier AI model clears 25% of policy tasks. Agents skip rules and ...
Gemini API Managed Agents: Hooks, Cron, and Free Tier
Google shipped its second major update to Gemini API Managed Agents in three weeks. The ...
Microsoft Agent Framework Harness: What It Actually Does
Microsoft shipped the Agent Framework Harness on July 22. Here's what the batteries-included runtime actually includes — and what you still have to ...
turbo-fieldfare: Gemma 4 26B in 2 GB RAM on Any Mac
turbo-fieldfare streams Gemma 4 26B experts from SSD, fitting a 26B model in 2 GB ...
OpenTelemetry Is a CNCF Grad. Your AI Agents Need It.
OpenTelemetry graduated CNCF. Here is what that means for AI agent observability, the GenAI semantic ...
Anthropic Mid-Conversation Tool Changes: No Cache Bust
Anthropic's beta lets you add or remove Claude tools mid-conversation without invalidating your prompt cache. ...
1,100 AI Employees Want to Slow AI. OpenAI and Anthropic Are Writing the Rules.
On July 28, 1,178 AI employees signed a letter asking the US to help slow ...