Measured over the last 30 days
- Models Tracked
- 24
- Avg Tokens / Second
- 12.40
- Avg Time to First Token (ms)
- 13407.52
- Last Updated
- Sep 2, 2026
All atlas-cloud models
24 tracked · fastest first| Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|
| KAT-Coder-Pro V2 | 35.80 | 31.60 | 41.40 | 1220.00 |
| DeepSeek V3.1 Terminus | 26.20 | 17.30 | 36.00 | 2380.00 |
| Qwen3 235B A22B Instruct 2507 | 25.00 | 22.10 | 27.40 | 1280.00 |
| DeepSeek V3.1 | 24.10 | 14.60 | 31.40 | 1960.00 |
| GLM 5.2 | 22.80 | 18.80 | 25.80 | 1920.00 |
| DeepSeek V3.2 Exp | 21.80 | 16.30 | 25.20 | 1850.00 |
| DeepSeek V3.2 | 17.60 | 13.70 | 23.10 | 2020.00 |
| DeepSeek V4 Flash 0423 | 16.90 | 15.30 | 18.30 | 3270.00 |
| GLM 5.1 | 14.40 | 2.47 | 23.20 | 8320.00 |
| MiniMax M2.5 | 13.10 | 4.09 | 28.40 | 8670.00 |
| DeepSeek V4 Flash 0731 | 12.20 | 1.03 | 25.50 | 20100.00 |
| MiMo-V2.5-Pro | 10.30 | 9.54 | 11.20 | 4860.00 |
| GLM 5 | 9.44 | 2.40 | 14.10 | 10900.00 |
| Kimi K2.7 Code | 7.80 | 3.63 | 10.00 | 9860.00 |
| Kimi K2.5 | 5.64 | 4.24 | 7.02 | 10500.00 |
| MiniMax M2.7 | 5.55 | 3.82 | 9.10 | 11700.00 |
| LongCat 2.0 | 5.33 | 1.53 | 7.82 | 16700.00 |
| Qwen3.5-35B-A3B | 5.32 | 4.64 | 5.87 | 11500.00 |
| Qwen3.5-122B-A10B | 5.31 | 4.33 | 5.87 | 12000.00 |
| Kimi K2.6 | 4.45 | 2.43 | 6.96 | 16000.00 |
| GLM 4.7 | 4.07 | 3.28 | 5.52 | 16100.00 |
| MiniMax M3 | 3.92 | 2.90 | 5.91 | 19100.00 |
| Qwen3.5-27B | 3.72 | 3.33 | 4.18 | 16800.00 |
| Qwen3.5 397B A17B | 3.49 | 2.88 | 3.98 | 18000.00 |
Frequently Asked Questions
Which atlas-cloud model is fastest?
Based on recent tests, KAT-Coder-Pro V2 shows the highest average throughput among tracked atlas-cloud models.
How many recent measurements feed this dashboard?
This provider summary aggregates 5434 individual prompts measured across 4510 monitoring runs over the past month.