Head to head
Mistral Medium 3.5 vs GPT-5.4
Mistral Medium 3.5: currentGPT-5.4: currentheavy tier
No verdict
Cost per solved task needs a published pass rate for both models. At least one has none, so it is reported as missing rather than estimated.
verified 2026-08-21 · prices from each provider's pricing page
Input $/M tokens
lower is better
Output $/M tokens
lower is better
Cost per solved task
needs a pass rate for both
| Mistral Medium 3.5 | GPT-5.4 | |
|---|---|---|
| Input $/Mtok | $1.5 | $2.5 |
| Output $/Mtok | $7.5 | $15 |
| Cached input $/Mtok | missing | $0.25 |
| Pass rate | missing | 59% |
| Pass rate source | missing | SEAL |
| Cost basis | missing | modelled |
| Cost per solved task | missing | $4.19 |
verified 2026-08-21 · heavy task tier; measured rows ignore the tier· Methodology· Mistral Medium 3.5· GPT-5.4