WideSearch — Qwen3.5 card; no explicit metric label; kept separate from item-f1-labeled variant on precision grounds
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Qwen3.8 Max Preview | Alibaba | 81.9 | 2 | 2026-08-23 |
| 2 | Claude Fable 5 | Anthropic | 81.2 | 1 | 2026-08-12 |
| 3 | Kimi K2.6 | Moonshot | 80.8 | 1 | 2026-08-23 |
| 4 | Kimi K2.5 | Moonshot | 79 | 1 | 2026-08-23 |
| 5 | Hy3-preview | Tencent | 76.4 | 1 | 2026-08-23 |
| 6 | Qwen3.7 Max | Alibaba | 75.2 | 1 | 2026-08-12 |
| 7 | Qwen3.6 Plus | Alibaba | 74.3 | 1 | 2026-08-23 |
| 8 | Qwen3.5 397B A17B | Alibaba | 74 | 1 | 2026-08-23 |
| 9 | Claude Opus 4.8 | Anthropic | 72.9 | 1 | 2026-08-12 |
| 10 | Qwen3.5 27B | Alibaba | 66.4 | 2 | 2026-08-23 |
| 11 | Qwen3.5 122B A10B | Alibaba | 60.5 | 1 | 2026-08-23 |
| 12 | Qwen3.6 35B A3B | Alibaba | 60.1 | 2 | 2026-08-23 |
| 13 | Qwen3.5 35B A3B | Alibaba | 59.1 | 2 | 2026-08-23 |
| 14 | Gemma 4 26B A4B | 38.3 | 1 | 2026-06-06 | |
| 15 | Gemma 4 31B | 35.2 | 1 | 2026-06-06 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.