AI Intel

OpenAI's Agents API has received updates, including the ability to spin up a browser with an agent via a single API call for computer use, and new Be…

MiniMax-M3.1 Flash Preview is now available on the Token Plan, designed to be faster and lighter for teams managing high-volume, latency-sensitive AI…

Anthropic has released Claude Sonnet 5.5, the second model in its 5.5 family, offering over 30% faster performance and up to 30% lower cost compared…

Anthropic has launched Claude Opus 5.5, the first model in its new 5.5 family, performing at the level of Claude Fable 5.1 for most tasks while costi…

Gemini Code Assist has seen multiple updates, including the General Availability of Gemini 2.5 Pro and Flash models, agent mode for individual users,…

New research introduces "Proxy Confidence" as a method to audit black-box LLM agents using a surrogate model's log-probabilities, addressing the chal…

SGAnalog has been introduced as an end-to-end circuit benchmark derived from open-source silicon tapeouts, aiming to evaluate whether AI models learn…

New research investigates how communication, even short messages, can improve decision-making in multi-agent LLM systems where agents have varying co…

New research proposes a method for training numerical intelligence in AI agents through auto-diagnosis and skill discovery, moving beyond mere code g…

The latest Claude Code GitHub release v2.1.290 introduces enhancements to tool usage reporting by adding serverToolUses and agentId to plugin events.