HMMT 2025 — Effort-tier variant (Kimi Thinking), very high scores (96.7-100) - distinct effort axis.
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | GPT-5 | OpenAI | 100 | 1 | 2026-05-15 |
| 2 | Kimi K2 (Reasoning) | Moonshot | 97.5 | 1 | 2026-05-15 |
| 3 | Grok 4 | SpaceXAI | 96.7 | 1 | 2026-05-15 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.