Head to head
Ministral 3 8B vs Qwen3.5 9B
On a blended workload of one million input tokens and three million output tokens, Ministral 3 8B is 1.5× cheaper than Qwen3.5 9B.
| Ministral 3 8B | Qwen3.5 9B | |
|---|---|---|
| Provider | Mistral | Together AI |
| Input / MTok | $0.15 | $0.17 |
| Output / MTok | $0.15 | $0.25 |
| Cache read / MTok | — | — |
| Context window | — | 262,144 |
| Blended 1M in + 3M out | $0.60 | $0.92 |
| Status | ga | ga |