WildAWM OpenCode - WebArena: leaderboard
WildAWM (2026): common OpenCode v1.17.7 agent harness, high reasoning effort, eight public web, claw and computer-use evaluations. Fine-tuned WildAWM-4B and WildAWM-35B-A3B remain distinct from Qwen baselines; cited OpenForgeRL and CUA-Gym rows use different harnesses and are not pooled into this protocol. Pass Rate (%); higher is better.
Metric: Pass Rate (%). Source: arxiv.org. Saturation forecast: Around January 2028. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 (High) | 75.66 | #12 (Claude Opus 5) |
| 2 | GPT-6 Astra (High) | 74.34 | #6 (GPT-6 Astra) |
| 3 | Claude Opus 4.8 (High) | 71.24 | #37 (Claude Opus 4.8) |
| 4 | GPT-5.5 (High) | 70.8 | #30 (GPT-5.5) |
| 5 | GPT-5.6 Sol (High) | 67.7 | #18 (GPT-5.6 Sol) |
| 6 | GLM-5.2 (High) | 67.26 | #73 (GLM-5.2) |
No result here: #4 Claude Opus 5.5, #7 Gemini 4 Argon, #8 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=wildawm-opencode-webarena · How It Works · Data refreshed daily, snapshot 2026-10-21.