Measured over the last 30 days
- Models Tracked
- 11
- Avg Tokens / Second
- 21.59
- Avg Time to First Token (ms)
- 7661.82
- Last Updated
- Oct 17, 2026
All nebius models
11 live · one row per model · fastest first · 1 folded behind their current lane| Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|
| GLM 5.2 | 65.80 | 65.80 | 65.80 | 930.00 |
| gpt-oss-120b | 38.10 | 12.60 | 56.50 | 1740.00 |
| Gemma 3 27B | 26.00 | 26.00 | 26.00 | 1130.00 |
| Qwen3 235B A22B Instruct 2507 | 25.20 | 20.10 | 33.60 | 870.00 |
| Muse Spark Open | 19.60 | 19.60 | 19.60 | 2840.00 |
| Qwen3 30B A3B Instruct 2507 | 19.60 | 15.70 | 23.40 | 1390.00 |
| Hermes 4 405B | 15.60 | 15.60 | 15.60 | 2280.00 |
| Nemotron 3 Nano 30B A3B | 11.90 | 11.90 | 11.90 | 4970.00 |
| DeepSeek V4 Pro 0813 | 8.39 | 6.75 | 10.00 | 7630.00 |
| DeepSeek V4 Flash 0731 | 5.77 | 5.77 | 5.77 | 10700.00 |
| GLM 5.1 | 1.49 | 0.88 | 2.09 | 49800.00 |
Frequently Asked Questions
Which nebius model is fastest?
Based on recent tests, GLM 5.2 shows the highest average throughput among tracked nebius models.
How many recent measurements feed this dashboard?
This provider summary aggregates 1379 individual prompts measured across 1268 monitoring runs over the past month.