Inference provider
Ionstream
3 models served. Speed, latency and price below are medians across this provider's catalogue, not per-endpoint measurements.
Models served
3
Median output speed
296 t/s
Median latency
306ms
Median price / 1M
$0.73
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
- V4.1 FlashDeepSeek39.5
- V4 Pro 0813DeepSeek36.0
- Qwen3.8 27BQwen33.7
View as table
| Model | Intelligence |
|---|---|
| V4.1 Flash (DeepSeek) | 39.5 |
| V4 Pro 0813 (DeepSeek) | 36.0 |
| Qwen3.8 27B (Qwen) | 33.7 |
Fastest output
Median tokens per second
- Qwen3.8 27BQwen409 t/s
- V4.1 FlashDeepSeek296 t/s
- V4 Pro 0813DeepSeek75 t/s
View as table
| Model | Tokens/s |
|---|---|
| Qwen3.8 27B (Qwen) | 409 |
| V4.1 Flash (DeepSeek) | 296 |
| V4 Pro 0813 (DeepSeek) | 75 |
Full catalogue
All 3 models on Ionstream
Filter by creator
3 of 3 models| V4.1 Flash DeepSeekopen | 39.5 | $0.07 |
| V4 Pro 0813 DeepSeekopen | 36.0 | $0.38 |
| Qwen3.8 27B Qwenopen | 33.7 | $0.19 |