Vendor
Moonshot AI
2 model versions evaluated ·5 runs in the corpus · rubric v15
Model versions
| Model | Meta-score | Appearances | Best at | Last seen |
|---|---|---|---|---|
| Moonshot AI Kimi Code k2.7 | 67.4 | 2 | Sprint-Plan Agent | 2026-07-10 |
| Moonshot AI Kimi k2.6 | 70.0 | 3 | Sprint-Review Agent | 2026-07-17 |
Runs (primary model)
| Run | SoW | Date | Composite |
|---|---|---|---|
| Kimi k2.6 | circles | Jul 2026 | 63.3 |
| Kimi Code k2.7 | circles | Jul 2026 | 61.1 |
| Kimi Code k2.7 | circles | Jul 2026 | 59.3 |
| Kimi k2.6 | circles | Jun 2026 | 53.1 |
| Kimi k2.6 | circles | Jul 2026 | 49.9 |
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.