Vendor
DeepSeek
2 model versions evaluated ·9 runs in the corpus · rubric v15 · DeepSeek site ↗
Model versions
| Model | Meta-score | Appearances | Best at | Last seen |
|---|---|---|---|---|
| DeepSeek Deepseek Flash v4 | 74.0 | 6 | Director | 2026-07-22 |
| DeepSeek Deepseek Pro v4 | 58.6 | 3 | Sprint-Plan Agent | 2026-07-05 |
Runs (primary model)
| Run | SoW | Date | Composite |
|---|---|---|---|
| Deepseek Flash v4 | circles | Jul 2026 | 70.1 |
| Deepseek Flash v4 | circles | Jul 2026 | 64.0 |
| Deepseek Flash v4 | circles | Jul 2026 | 56.8 |
| Deepseek Flash v4 | circles | Jul 2026 | 56.0 |
| Deepseek Pro v4 | circles | Jul 2026 | 55.9 |
| Deepseek Flash v4 | the2048 | Jun 2026 | 55.6 |
| Deepseek Pro v4 | circles | Jun 2026 | 53.9 |
| Deepseek Pro v4 | circles | Jul 2026 | 51.9 |
| Deepseek Flash v4 | solarSystem | Jun 2026 | 50.2 |
Every run on Favur Evals is scored by the same deterministic engine, on the same Statements of Work. The benchmark is self-funded — no vendor sponsorship, credits, or grants.