SkillsBench — Qwen3.6-card agentic-skills benchmark, averaged-over-5-runs metric.
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Qwen3.6 27B | Alibaba | 48.2 | 1 | 2026-06-15 |
| 2 | Claude Opus 4.5 | Anthropic | 45.3 | 1 | 2026-06-15 |
| 3 | Qwen3.5 397B A17B | Alibaba | 30 | 1 | 2026-06-15 |
| 4 | Qwen3.6 35B A3B | Alibaba | 28.7 | 2 | 2026-06-15 |
| 5 | Qwen3.5 27B | Alibaba | 27.2 | 2 | 2026-06-15 |
| 6 | Gemma 4 31B | 23.6 | 2 | 2026-06-15 | |
| 7 | Gemma 4 26B A4B | 12.3 | 1 | 2026-05-03 | |
| 8 | Qwen3.5 35B A3B | Alibaba | 4.4 | 1 | 2026-05-03 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.