STT-Arena — leaderboard
Metric: Overall (self-reported). Source: benchmarklist.com. 23 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 | 35.39 |
| 2 | GPT-5.4 | 34.51 |
| 3 | Gemini 3.1 Pro (Preview) | 33.19 |
| 4 | Claude Sonnet 4.6 | 32.74 |
| 5 | DeepSeek V3.2 | 32.16 |
| 6 | Qwen 3.6 Plus | 31.42 |
| 7 | Qwen 3.5 397B A17B | 28.78 |
| 8 | GLM-5.1 | 27.31 |
| 9 | DeepSeek V4 Pro | 26.29 |
| 10 | GLM-5 | 24.23 |
| 11 | MiniMax-M2.5 | 21.88 |
| 12 | MiniMax-M2.7 | 21.29 |
| 13 | Gemini 2.5 Flash | 20.11 |
| 14 | Qwen 3.5 9B | 19.53 |
| 15 | Llama 3.3 70B Instruct | 17.18 |
Interactive version: theaggregate.ai/benchmark?slug=stt-arena · How the rankings work · Data refreshed daily, snapshot 2026-07-22.