Measured over the last 30 days
- Models Tracked
- 10
- Avg Tokens / Second
- 13.68
- Avg Time to First Token (ms)
- 19045.00
- Last Updated
- Oct 17, 2026
All digitalocean models
10 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 | 34.60 | 23.00 | 46.20 | 1120.00 |
| DeepSeek V3.2 | 26.10 | 16.90 | 44.80 | 950.00 |
| Llama 4 Maverick | 24.70 | 10.40 | 39.10 | 650.00 |
| DeepSeek V4 Flash 0731 | 15.90 | 11.40 | 20.30 | 2050.00 |
| DeepSeek V4 Flash 0423 | 13.50 | 2.93 | 31.90 | 1350.00 |
| gpt-oss-120b | 10.50 | 4.97 | 18.40 | 5330.00 |
| MiniMax M2.5 | 6.32 | 3.69 | 10.80 | 10500.00 |
| Kimi K2.6 | 3.04 | 1.47 | 6.34 | 23200.00 |
| Qwen3.5 397B A17B | 1.47 | 0.96 | 1.87 | 51000.00 |
| DeepSeek V4.1 Flash | 0.66 | 0.66 | 0.66 | 94300.00 |
Frequently Asked Questions
Which digitalocean model is fastest?
Based on recent tests, GLM 5.2 shows the highest average throughput among tracked digitalocean models.
How many recent measurements feed this dashboard?
This provider summary aggregates 2209 individual prompts measured across 2016 monitoring runs over the past month.