Inference provider
Ionstream
4 models served. Speed, latency and price below are medians across this provider's catalogue, not per-endpoint measurements.
Models served
4
Median output speed
100 t/s
Median latency
352ms
Median price / 1M
$0.214
Catalogue
Strongest and fastest on this provider
What the catalogue looks like at the top end, on the two axes that usually decide the choice.
Highest Intelligence Index
Among models this provider serves
- V4 Flash 0731DeepSeek51.8
- V4 ProDeepSeek45.3
- Qwen3 Coder NextQwen21.3
View as table
| Model | Intelligence |
|---|---|
| V4 Flash 0731 (DeepSeek) | 51.8 |
| V4 Pro (DeepSeek) | 45.3 |
| Qwen3 Coder Next (Qwen) | 21.3 |
Fastest output
Median tokens per second
- V4 Flash 0731DeepSeek148 t/s
- Qwen3 Coder NextQwen129 t/s
- V4 ProDeepSeek71 t/s
- Gemma 4 26B A4B Google71 t/s
View as table
| Model | Tokens/s |
|---|---|
| V4 Flash 0731 (DeepSeek) | 148 |
| Qwen3 Coder Next (Qwen) | 129 |
| V4 Pro (DeepSeek) | 71 |
| Gemma 4 26B A4B (Google) | 71 |
Full catalogue
All 4 models on Ionstream
Filter by creator
4 of 4 models| V4 Flash 0731 DeepSeekopen | 51.8 | $0.02 |
| V4 Pro DeepSeekopen | 45.3 | $0.09 |
| Qwen3 Coder Next Qwenopen | 21.3 | $0.03 |
| Gemma 4 26B A4B Googleopen | — | $0.03 |