SWE-bench Pro (Opus 5 & Mythos 5.1 System Cards): leaderboard
SWE-bench Pro, a harder software-engineering benchmark of long-horizon tasks from professional repositories, in Anthropic's own runs; percentage of tasks resolved (higher is better). From Table 8.1.A of the Claude Fable 5.1 & Claude Mythos 5.1 System Card (September 2026), matching the Opus 5 card; comparator cells already on SWE-bench Pro (Anthropic Scaffold) are omitted.
Metric: Resolved (%). Source: www-cdn.anthropic.com. Saturation forecast: Around February 2027. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 (Max) | 89.9 | #3 (Claude Opus 5.5) |
| 2 | Claude Fable 5.1 (Max) | 81.2 | #10 (Claude Fable 5.1) |
| 3 | Claude Opus 5 (Max) | 79.2 | #16 (Claude Opus 5) |
No result here: #6 GPT-6 Astra, #7 GPT-6.1 Sol, #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=swe-bench-pro-opus-5-mythos-5-1-system-cards · How It Works · Data refreshed daily, snapshot 2026-11-05.