STT-Arena - Hard: leaderboard
Metric: Pass@1 (%) on the 59 hard tasks (long-horizon cascading constraints requiring global replanning); a task passes when every rule-based check function holds in the final state; the model acts in an executable stateful environment whose state changes under scheduled spatio-temporal triggers, with a passive Qwen3.5-397B user simulator and at most 50 turns; mean of three runs at temperature 0.7; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 23 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 | 24.29 |
| 2 | Claude Opus 4.6 | 23.73 |
| 3 | Qwen 3.6 Plus | 23.16 |
| 4 | Gemini 3.1 Pro (Preview) | 21.47 |
| 5 | Claude Sonnet 4.6 | 19.77 |
| 6 | DeepSeek V3.2 | 15.82 |
| 7 | DeepSeek V4 Pro | 15.25 |
| 8 | GLM-5.1 | 15.25 |
| 9 | Kimi K2.5 | 14.12 |
| 10 | MiniMax-M2.7 | 13.56 |
| 11 | Gemini 2.5 Flash | 12.43 |
| 12 | MiniMax-M2.5 | 12.43 |
| 13 | Qwen 3.5 9B | 11.3 |
| 14 | Qwen 3 8B | 11.3 |
| 15 | Llama 3.3 70B | 11.3 |
Interactive version: theaggregate.ai/benchmark?slug=stt-arena-hard · How It Works · Data refreshed daily, snapshot 2026-10-07.