Anthropic releases advanced tool use in beta

advanced tool use beta — Tool Search Tool, Programmatic Tool Calling

Why it matters

Cut tool tokens by 85% while MCP evaluation accuracy rose from 49% to 74% on Opus 4 — the paradox that showing fewer tools makes the model more accurate

Source — Primary sources first; secondary reports are labelled “Press”

Officialanthropic.comhttps://www.anthropic.com/engineering/advanced-tool-use

Added in Week 35, 2026 · See that week

Same week

vLLM project announces vLLM-OmniOfficialLiquid AI publishes the LFM2 technical reportPaperEffective harnesses for long-running agentsOfficialModel Context Protocol revises its specOfficial

Nearby on the same axis

Model Context Protocol revises its specOfficialAnthropic, OpenAI and MCP-UI propose the MCP Apps extension (SEP-1865)OfficialAnthropic, Block and OpenAI donate MCP to the Linux Foundation's AAIFOfficialAnthropic publishes Agent Skills as an open standard (agentskills.io)Press