STT-Arena - Medium: leaderboard
Medium-difficulty tasks of STT-Arena (2026), a tool-use benchmark in executable stateful environments whose conditions change in time and space during the task, with a simulated user. Pass@1 in percent (every rule-based check must hold in the final state), mean of three runs; higher is better.
Metric: Pass@1 (%) on the 67 medium tasks; 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 January 2028. 21 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.6 | 31.34 | #114 |
| 2 | Gemini 3.1 Pro (Preview) | 30.35 | #102 |
| 3 | Claude Sonnet 4.6 | 29.85 | #179 |
| 4 | GPT-5.4 | 29.85 | #138 |
| 5 | DeepSeek V3.2 | 27.86 | #425 |
| 6 | Qwen 3.6 Plus | 24.38 | #245 |
| 7 | GLM-5.1 | 23.38 | #200 |
| 8 | GLM-5 | 21.39 | #276 |
| 9 | DeepSeek V4 Pro | 20.4 | #201 |
| 10 | STT-Agent-4B (SFT+RL) | 19.9 | #324 |
| 11 | STT-Agent-4B (SFT) | 17.41 | #424 |
| 12 | MiniMax-M2.7 | 15.92 | #539 |
| 13 | MiniMax-M2.5 | 15.92 | #600 |
| 14 | Gemini 2.5 Flash | 13.43 | #466 |
| 15 | Kimi K2.5 | 13.43 | #272 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=stt-arena-medium · How It Works · Data refreshed daily, snapshot 2026-11-22.