Measured over the last 30 days
- Models Tracked
- 17
- Avg Tokens / Second
- 18.12
- Avg Time to First Token (ms)
- 4446.63
- Last Updated
- Sep 1, 2026
| Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|
| Claude Haiku 4.5 | 38.80 | 31.80 | 45.60 | 940.00 |
| GPT-3.5 Turbo (older v0613) | 38.10 | 27.00 | 50.40 | 1380.00 |
| GPT-4o (2024-08-06) | 36.30 | 15.90 | 51.20 | 1590.00 |
| GPT-4o | 35.30 | 27.30 | 50.20 | 1570.00 |
| GPT-4o-mini | 28.90 | 23.00 | 35.90 | 1870.00 |
| GPT-5.1-Codex | 27.40 | 7.39 | 45.70 | 2500.00 |
| Claude Sonnet 5 | 26.60 | 20.00 | 29.60 | 1120.00 |
| DeepSeek V4 Pro 0423 | 26.50 | 22.60 | 29.50 | 1890.00 |
| Claude Sonnet 4.6 | 23.60 | 22.00 | 25.70 | 1480.00 |
| DeepSeek V4 Flash 0423 | 23.60 | 8.32 | 36.30 | 3520.00 |
| Claude Opus 4.5 | 23.10 | 19.70 | 25.90 | 1330.00 |
| Claude Opus 4.7 | 21.50 | 19.00 | 22.70 | 1440.00 |
| Claude Opus 4.8 | 21.40 | 11.10 | 24.10 | 1770.00 |
| GPT-5.2-Codex | 19.70 | 9.67 | 26.70 | 3390.00 |
| Claude Sonnet 4.5 | 19.20 | 18.00 | 20.50 | 1660.00 |
| Claude Opus 4.6 | 13.90 | 6.19 | 16.80 | 3890.00 |
| GPT-4o (2024-05-13) | 13.00 | 10.30 | 36.30 | 5380.00 |
Frequently Asked Questions
Which azure model is fastest?
Based on recent tests, Claude Haiku 4.5 shows the highest average throughput among tracked azure models.
How many recent measurements feed this dashboard?
This provider summary aggregates 8749 individual prompts measured across 4431 monitoring runs over the past month.