| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | LFM2.5-8B-A1B | Liquid AI | 88.1 | 1 | 2026-08-09 |
| 2 | Qwen3.5 4B | Alibaba | 87.7 | 1 | 2026-08-09 |
| 3 | Gemma 4 E4B | 26.8 | 1 | 2026-08-09 | |
| 4 | Gemma 4 E2B | 22.4 | 1 | 2026-08-09 | |
| 5 | Qwen3 30B A3B | Alibaba | 21.9 | 1 | 2026-08-09 |
| 6 | Granite 4.0 H Tiny | IBM | 16.7 | 1 | 2026-08-09 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.