Head to head
GPT-5 nano vs Devstral Small 2
On a blended workload of one million input tokens and three million output tokens, Devstral Small 2 is 1.3× cheaper than GPT-5 nano.
| GPT-5 nano | Devstral Small 2 | |
|---|---|---|
| Provider | OpenAI | Mistral |
| Input / MTok | $0.05 | $0.1 |
| Output / MTok | $0.4 | $0.3 |
| Cache read / MTok | $0.005 | — |
| Context window | — | — |
| Blended 1M in + 3M out | $1.25 | $1.00 |
| Status | ga | ga |