# WeaveBench Leaderboard

> Last updated AUG 31, 2026

Tracking reported computer-use agent performance on WeaveBench, combining the official benchmark leaderboard with additional results found by Yutori Scout.

Official results come from the WeaveBench leaderboard. The official leaderboard has not been updated since Aug 22, 2026. We’ve set up a [Scout](<https://yutori.com/scouting>) to monitor for additional results and catch results it has missed.

## Reported results

| Model | Harness | Overall score | PR |
| --- | --- | --- | --- |
| [Yutori Navigator n2](<https://yutori.com/blog/introducing-n2>)<sup>1</sup> | Yutori Agent | 70.3% | 54.75%<sup>2</sup> |
| Seed 2.1 turbo | Claude Code | 57.1% | 28.9% |
| Seed 2.1 pro | Claude Code | 55.1% | 30.7% |
| Claude Opus 4.7 | Claude Code | 53.2% | 41.2% |
| Claude Opus 4.7 | Hermes Agent | 51.6% | 28.1% |
| GPT-5.5 | Codex CLI | 49.9% | 35.1% |
| Claude Opus 4.7 | OpenClaw | 48.2% | 35.1% |
| GPT-5.5 | OpenClaw | 46.6% | 33.3% |
| GPT-5.5 | Hermes Agent | 46.6% | 31.6% |
| GPT-5.4 | OpenClaw | 46.5% | 22.8% |
| GPT-5.3-codex | OpenClaw | 45.6% | 18.4% |
| Claude Opus 4.7 | Codex CLI | 37.8% | 13.2% |
| GPT-5.2-codex | OpenClaw | 32.1% | 6.1% |
| Qwen3.5-397B-A17B | OpenClaw | 31.8% | 0.9% |
| GPT-5.5 | Claude Code | 29.9% | 14.9% |
| GPT-5.1-codex | OpenClaw | 22.6% | 1.8% |
| Gemini 3.1 Pro | OpenClaw | 22.3% | 1.8% |
| Qwen3-VL-8B-Think | OpenClaw | 9.2% | 0.9% |
| GUI-Owl-1.5-32B | OpenClaw | 6.5% | 0.0% |

<sup>1</sup> Navigator n2’s Overall score is reported in the [launch post](<https://yutori.com/blog/introducing-n2>).

<sup>2</sup> Navigator n2’s 54.75% PR and Yutori Agent harness are first-party Yutori data supplied for this leaderboard. The PR was obtained with a 500-step limit.

## About the benchmark

WeaveBench evaluates 114 long-horizon tasks across 8 work domains. Every task requires an agent to interleave GUI observation or action with CLI or code execution in one trajectory inside a real Ubuntu sandbox.

## Sources

- [Paper (arXiv 2606.09426)](<https://arxiv.org/abs/2606.09426>)
- [Code (GitHub)](<https://github.com/weavebench/WeaveBench>)
- [Dataset (Hugging Face)](<https://huggingface.co/datasets/wanlilll/WeaveBench>)
- [Official leaderboard](<https://weavebench.github.io/#leaderboard>)
- [Navigator n2 launch post](<https://yutori.com/blog/introducing-n2>)

Official source revision (SHA-256): `bd125453d36a2ec286cf1106c60f578bd4e83dbbc98bc7bcb78091f648fe5245`