Measured over the last 30 days
- Models Tracked
- 18
- Avg Tokens / Second
- 11.34
- Avg Time to First Token (ms)
- 8660.56
- Last Updated
- Oct 17, 2026
All gmicloud models
18 live · one row per model · fastest first · 5 folded behind their current lane| Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|
| MiniMax M3 | 23.20 | 22.50 | 23.70 | 1920.00 |
| DeepSeek V3 0324 | 22.10 | 17.60 | 26.60 | 1880.00 |
| Qwen3 235B A22B Instruct 2507 | 19.60 | 6.73 | 27.60 | 3320.00 |
| GLM 5.1 | 18.40 | 15.50 | 20.30 | 2660.00 |
| DeepSeek V4 Flash 0731 | 16.20 | 3.63 | 27.40 | 4530.00 |
| GLM 5.2 | 15.10 | 15.10 | 15.10 | 2810.00 |
| DeepSeek V4 Flash 0423 | 14.90 | 4.18 | 26.30 | 4920.00 |
| MiMo-V2.5-Pro | 13.00 | 13.00 | 13.00 | 3130.00 |
| MiMo-V2.6-Flash | 12.30 | 10.80 | 13.90 | 3540.00 |
| DeepSeek V3.2 | 8.11 | 2.16 | 17.40 | 10800.00 |
| Kimi K2.7 Code | 7.93 | 6.14 | 9.56 | 7170.00 |
| MiniMax M2.5 | 7.67 | 5.07 | 10.80 | 9990.00 |
| MiMo-V2.5 | 6.51 | 6.51 | 6.51 | 7620.00 |
| MiMo-V2.6-Pro | 5.30 | 2.25 | 8.36 | 15400.00 |
| MiniMax M2.7 | 3.87 | 2.09 | 4.76 | 20000.00 |
| DeepSeek V4.1 Flash | 3.71 | 3.41 | 4.00 | 17100.00 |
| Kimi K2.6 | 3.56 | 3.56 | 3.56 | 16500.00 |
| GLM 5 | 2.71 | 2.71 | 2.71 | 22600.00 |
Frequently Asked Questions
Which gmicloud model is fastest?
Based on recent tests, MiniMax M3 shows the highest average throughput among tracked gmicloud models.
How many recent measurements feed this dashboard?
This provider summary aggregates 2220 individual prompts measured across 2016 monitoring runs over the past month.