Longform Writing eval (Kimi K2 Thinking system card) — Identical stats to Longform Writing no tools -- same report, likely no-tools condition unlabeled.
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4.5 | Anthropic | 79.8 | 2 | 2026-06-12 |
| 2 | Kimi K2 (Reasoning) | Moonshot | 73.8 | 3 | 2026-06-12 |
| 3 | DeepSeek-V3.2 | DeepSeek | 72.5 | 2 | 2026-06-12 |
| 4 | GPT-5 | OpenAI | 71.4 | 1 | 2026-06-12 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.