Head to head
Qwen2.5 7B Instruct Turbo vs Devstral Small 2
On a blended workload of one million input tokens and three million output tokens, Devstral Small 2 is 1.2× cheaper than Qwen2.5 7B Instruct Turbo.
| Qwen2.5 7B Instruct Turbo | Devstral Small 2 | |
|---|---|---|
| Provider | Together AI | Mistral |
| Input / MTok | $0.3 | $0.1 |
| Output / MTok | $0.3 | $0.3 |
| Cache read / MTok | — | — |
| Context window | 32,768 | — |
| Blended 1M in + 3M out | $1.20 | $1.00 |
| Status | ga | ga |