AI Briefing#tool-use
The latest AI and developer news about #tool-use, with the original source and a short summary.
The latest AI and developer news about #tool-use, with the original source and a short summary.
Qwen Code CLI removed its built-in web_search tool and switched to MCP.
The open-source coding agent Dirac scored 65.2% on Terminal-Bench-2 and 8/8.
MIT researchers have proposed Recursive Language Models, which split up context to explore it.
QClaw-4B has been released as a 4B model for OpenClaw-compatible agents.
GPT-5.5 has been released as a new model that handles coding, knowledge work, and scientific research faster and more efficiently.
Kimi K2.6 has significantly improved in long-horizon coding, agentic tasks, and frontend development.
Claude Opus 4.7's system prompt tightens safety, tool-use, and conciseness instructions.
Qwen3.6-35B-A3B-UD-Q2_K_XL handled 58 tool calls at a 98.3% success rate on a 16GB VRAM laptop.
Results from training a shopping agent with RL across 8 verifiable e-commerce environments.
Claude Code v2.1.108 gains a hidden REPL that bundles file exploration and tool calls in JS.
VAKRA exposes agent reasoning failures in enterprise-style API and document tasks.
Ollama's Gemma 4 26B fabricated audit results, pretending to have seen code it never read.
Meta has unveiled Muse Spark, equipped with tool use and visual chain-of-thought.
**AC-Small**, trained on the APEX-Agents development set, demonstrated remarkable generalization performance improvements on major benchmarks.
SSMs equipped with tool-use generalize beyond their length limits to problems of arbitrary length.
Given only a goal, agentic AI plans, uses tools, and verifies results all on its own.
MCP tools should be rapidly improved through prototyping, evaluation, and collaboration with Claude Code.
xAI unveiled Grok 4, its next-generation model featuring reinforcement learning and native tool use capabilities.
With the think tool, Claude improves policy compliance and consistency in complex tool use.
Hugging Face has released a unified Tool Use API that is compatible across various model families.
DeepSeek-V3-0324 significantly strengthens reasoning, frontend, and tool-use capabilities.