Head to head
Magistral Medium vs GPT-5.4 mini
On a blended workload of one million input tokens and three million output tokens, GPT-5.4 mini is 1.2× cheaper than Magistral Medium.
| Magistral Medium | GPT-5.4 mini | |
|---|---|---|
| Provider | Mistral | OpenAI |
| Input / MTok | $2 | $0.75 |
| Output / MTok | $5 | $4.5 |
| Cache read / MTok | — | $0.075 |
| Context window | — | — |
| Blended 1M in + 3M out | $17.00 | $14.25 |
| Status | ga | ga |