Institution

Infinigence AI

AI infrastructure and systems company co-authoring research on coding agents and SWE-style evaluation.

Claw-SWE-Bench: Why Coding Agent Harnesses Matter

Claw-SWE-Bench evaluates OpenClaw-style coding-agent harnesses on 350 GitHub issue tasks. OpenClaw jumps from 19.1% to 73.4% Pass@1 with a full adapter.