Measured over the last 30 days
- Models Tracked
- 21
- Avg Tokens / Second
- 14.35
- Avg Time to First Token (ms)
- 8347.14
- Last Updated
- Oct 14, 2026
All streamlake models
21 live · one row per model · fastest first · 6 folded behind their current lane| Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|
| Qwen3 30B A3B Instruct 2507 | 42.20 | 42.20 | 42.20 | 990.00 |
| Qwen3 Coder Next | 40.40 | 34.00 | 46.80 | 1200.00 |
| DeepSeek V3.1 Terminus | 30.10 | 28.40 | 31.90 | 1480.00 |
| MiMo-V2.5-Pro | 27.00 | 27.00 | 27.00 | 1550.00 |
| DeepSeek V3 | 25.50 | 25.50 | 25.50 | 1890.00 |
| Qwen3 235B A22B Instruct 2507 | 18.70 | 8.71 | 25.10 | 2640.00 |
| GLM 5.2 | 18.20 | 17.70 | 18.70 | 2340.00 |
| DeepSeek V3.2 | 15.50 | 15.50 | 15.50 | 2070.00 |
| MiniMax M3 | 13.80 | 9.00 | 18.10 | 4140.00 |
| MiniMax M2.5 | 9.65 | 8.42 | 10.90 | 6290.00 |
| Kimi K2.7 Code | 9.56 | 3.93 | 17.30 | 9040.00 |
| DeepSeek V4 Pro 0423 | 7.37 | 3.25 | 11.40 | 8640.00 |
| DeepSeek V4 Flash 0423 | 6.67 | 5.50 | 7.76 | 8890.00 |
| MiMo-V2.5 | 6.63 | 6.63 | 6.63 | 7630.00 |
| R1 0528 | 5.69 | 4.80 | 6.57 | 10600.00 |
| Qwen3.5 397B A17B | 5.36 | 0.61 | 9.60 | 12300.00 |
| DeepSeek V4 Flash 0731 | 5.35 | 4.87 | 5.83 | 11000.00 |
| Kimi K2.6 | 4.78 | 3.26 | 6.62 | 12900.00 |
| DeepSeek V4.1 Flash | 4.14 | 3.90 | 4.38 | 14800.00 |
| GLM 5 | 2.88 | 2.88 | 2.88 | 20800.00 |
| GLM 5.1 | 1.90 | 1.43 | 2.73 | 34100.00 |
Frequently Asked Questions
Which streamlake model is fastest?
Based on recent tests, Qwen3 30B A3B Instruct 2507 shows the highest average throughput among tracked streamlake models.
How many recent measurements feed this dashboard?
This provider summary aggregates 2326 individual prompts measured across 2084 monitoring runs over the past month.