Inference provider
Friendli
5 models served. Speed, latency and price below are medians across this provider's catalogue, not per-endpoint measurements.
Models served
5
Median output speed
100 t/s
Median latency
549ms
Median price / 1M
$0.39
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
View as table
| Model | Intelligence |
|---|---|
| GLM 5.2 (Z.ai) | 52.6 |
| GLM 5.1 (Z.ai) | 41.0 |
| M2.5 (MiniMax) | 34.5 |
| V3.2 (DeepSeek) | 25.1 |
Fastest output
Median tokens per second
- Gemma 4 31BGoogle232 t/s
- GLM 5.2Z.ai142 t/s
- V3.2DeepSeek100 t/s
- M2.5MiniMax91 t/s
- GLM 5.1Z.ai67 t/s
View as table
| Model | Tokens/s |
|---|---|
| Gemma 4 31B (Google) | 232 |
| GLM 5.2 (Z.ai) | 142 |
| V3.2 (DeepSeek) | 100 |
| M2.5 (MiniMax) | 91 |
| GLM 5.1 (Z.ai) | 67 |
Full catalogue
All 5 models on Friendli
Filter by creator
5 of 5 models| GLM 5.2 Z.aiopen | 52.6 | $0.23 |
| GLM 5.1 Z.aiopen | 41.0 | $0.29 |
| M2.5 MiniMaxopen | 34.5 | $0.08 |
| V3.2 DeepSeekopen | 25.1 | $0.05 |
| Gemma 4 31B Googleopen | — | $0.03 |