Arena-Hard-Auto (Creative Writing subset) — Explicit category subset of Arena-Hard, distinct domain from Hard Prompt subset
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Gemini 3 Pro | 93.6 | 2 | 2026-06-13 | |
| 2 | GPT-5 | OpenAI | 92.2 | 1 | 2026-06-06 |
| 3 | DeepSeek-V3.2 | DeepSeek | 88.8 | 2 | 2026-06-13 |
| 4 | MiMo V2 Flash | Xiaomi | 86.2 | 2 | 2026-06-13 |
| 5 | Kimi K2 (Reasoning) | Moonshot | 80.1 | 2 | 2026-06-13 |
| 6 | Claude Sonnet 4.5 | Anthropic | 76.7 | 1 | 2026-06-06 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.