Artificial Analysis Coding Agent Index v2.0: Claude Opus 5.5 in Claude Code Leads at 61.7
jffwng · x · 2026-10-23
Artificial Analysis launched Coding Agent Index v2.0, now spanning 290 tasks across software engineering, terminal use, and repo Q&A weighted equally, evaluating models with their agent harnesses.
- FrontierCode v1.1 (from Cognition, private test set of 100 real-repo tasks graded by maintainer-merge-worthiness) replaces DeepSWE; SWE-Atlas-QnA grading revised
- Claude Opus 5.5 in Claude Code leads at 61.7
- Muse Wonder Full + Muse Code lands third per the reposter
More from coding & agent
- Mubit trains small models from your own agent API traffic to cut cost and latency — rohanpaul_ai · 2026-10-24
- Heavy AI user: marketed automations are fragile, exhausting, and dumber than we admit — mrks86 · 2026-10-24
- Google's playbook: evaluating AI agents beyond the vibe check — bibryam · 2026-10-24
- Running CI weekly on dotfiles setup.sh inside a fresh Ubuntu container to catch breakage — 4310sy · 2026-10-24
- Codex CLI's hidden remote-control feature praised as better than Claude Code's — antirez · 2026-10-24
- PWAs can't use foldable screens on the Duo yet, but two Chrome flags unlock the APIs — mitsuhiko · 2026-10-24