Measured over the last 30 days
- Models Tracked
- 13
- Avg Tokens / Second
- 25.02
- Avg Time to First Token (ms)
- 3249.23
- Last Updated
- Nov 8, 2026
All together models
13 live · one row per model · fastest first · 2 folded behind their current lane| Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|
| GLM 5.2 | 55.60 | 33.20 | 76.60 | 780.00 |
| Beam 1.0 501B A23B | 43.90 | 30.10 | 61.70 | 1150.00 |
| gpt-oss-120b | 35.70 | 32.70 | 41.30 | 1350.00 |
| Llama 3.3 70B Instruct | 33.80 | 22.70 | 61.00 | 1270.00 |
| Glimpse | 23.80 | 6.83 | 45.40 | 2600.00 |
| Muse Spark Open | 21.90 | 16.70 | 24.60 | 2650.00 |
| DeepSeek V4.1 Flash 1017 | 20.50 | 6.47 | 40.40 | 3490.00 |
| Qwen3.8 2.4T A95B | 19.00 | 12.50 | 25.50 | 3360.00 |
| DeepSeek V4 Flash 0731 | 18.10 | 18.10 | 18.10 | 2570.00 |
| MiniMax M3 | 17.90 | 3.97 | 28.10 | 4860.00 |
| Nemotron 3.5 Super VL | 16.90 | 6.92 | 28.80 | 4210.00 |
| Muse Glimmer 30B | 11.20 | 11.20 | 11.20 | 5150.00 |
| Qwen3.5-9B | 7.00 | 6.56 | 7.44 | 8800.00 |
Frequently Asked Questions
Which together model is fastest?
Based on recent tests, GLM 5.2 shows the highest average throughput among tracked together models.
How many recent measurements feed this dashboard?
This provider summary aggregates 3044 individual prompts measured across 2581 monitoring runs over the past month.