Vendor
Anthropic
2 model versions evaluated ·7 runs in the corpus · rubric v15 · Anthropic site ↗
Model versions
| Model | Meta-score | Appearances | Best at | Last seen |
|---|---|---|---|---|
| Anthropic Claude Sonnet 4.6 | 43.2 | 2 | Scout Agent | 2026-07-03 |
| Anthropic Claude Sonnet 5 | 66.9 | 5 | Sprint-Review Agent | 2026-07-23 |
Runs (primary model)
| Run | SoW | Date | Composite |
|---|---|---|---|
| Claude Sonnet 5 | circles | Jul 2026 | 70.8 |
| Claude Sonnet 5 + Gemini Flash Lite | solarSystem | Jul 2026 | 58.3 |
| Claude Sonnet 5 | circles | Jul 2026 | 57.9 |
| Claude Sonnet 5 | circles | Jul 2026 | 47.8 |
| Claude Sonnet 5 | circles | Jul 2026 | 46.6 |
| Claude Sonnet 4.6 | circles | Jul 2026 | 44.4 |
| Claude Sonnet 4.6 | circles | Jun 2026 | 42.3 |
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.