Composio benchmark finds similar success rates but wide speed gaps across coding harnesses
Teknium · x · 2026-07-29
A Composio comparison shows that three coding harnesses ended up with similar success rates, but very different speed and token efficiency.
- Kimi Code: 22/28 tasks
- Hermes: 21/28 tasks
- Claude Code: 20/28 tasks
The split was in runtime:
- Hermes: median 179s per task
- Kimi Code: 297s
- Claude Code: 348s
In short, the fastest harness and the most token-efficient harness were not the same one.
Related event: Composio Test: Agent Frameworks Show 30x Token Gap(7 posts)→
More from coding & agent
- 'LLM code is sloppy' rebutted with real ffmpeg code: identical if/else branches in a top library — kingslyj · 2026-10-16
- Office-editing tool calls need file, worktree and unit: lessons from dsh-univer-office — FunDistribution6537 · 2026-10-16
- Architect MD workflow: Gemini one-shot outputs that beat Opus, via 5-lens plan critique — oikk01 · 2026-10-16
- Prove a coding agent actually ran the tests: check artifacts, not approvals — ComprehensiveShake76 · 2026-10-16
- Building an acceptance harness for AI spreadsheet edits: allowlists, recalc checks, idempotency — ukdicha-modakkk · 2026-10-16
- Agent actions need checks outside the model: NVIDIA demos lean on guardrails amid wave of real agent incidents — Creamy-And-Crowded · 2026-10-16