ClawArena (Claude Code) - Task Completion Rate: leaderboard

ClawArena (2026) tests agents as persistent assistants in evolving information environments: 12 multi-turn scenarios with conflicting sources, staged updates and implicit preferences, checked by multiple-choice and executable shell questions, here with the agent run inside the Claude Code harness. Task completion rate in percent; higher is better.

Metric: Task completion rate (%) on ClawArena (12 multi-turn scenarios, 337 evaluation rounds; per-round score macro-averaged over scenarios, multi-choice and executable checks); agent run in the Claude Code framework; higher is better. Source: arxiv.org. Saturation forecast: Around June 2027. 6 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Claude Opus 4.776.13#85
2Claude Sonnet 4.673.36#173
3Claude Haiku 4.572.29#583
4Kimi K2.566.77#266
5GPT-5.559.31#54
6GPT-5.155.19#240

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=clawarena-claude-code-task-completion-rate · How It Works · Data refreshed daily, snapshot 2026-11-21.