Head to head
Mistral Medium 3 vs GPT-4.1 mini
Mistral Medium 3 from Mistral AI against GPT-4.1 mini from OpenAI — specification, price and every benchmark both makers have published, in one table.
Benchmarks
GPT-4.1 mini leads
GPT-4.1 mini wins 2 of the 2 benchmarks both models report, Mistral Medium 3 wins 0, consistently. The average gap across shared tests is 4.6 points.
Price
GPT-4.1 mini is cheaper
On a 3:1 input-to-output mix, GPT-4.1 mini costs $0.70 per million tokens against $0.80 for Mistral Medium 3 — a small difference.
What actually differs
- GPT-4.1 mini takes 1M tokens of context against 131K — 8.0× more room for long documents or a large codebase.
Scorecard
Which is better at what
Maths, coding, reasoning and the rest — one line each, averaged over the benchmarks both models actually report.
| Category | Mistral Medium 3 | GPT-4.1 mini | Better at this |
|---|---|---|---|
| ReasoningGPQA Diamond | 62.4% | 65.0% | GPT-4.1 mini+2.6 |
| MathsNeither model reports this | — | — | Not comparable |
| CodingOnly one model reports this | — | — | Not comparable |
| KnowledgeOnly one model reports this | — | — | Not comparable |
| MultimodalMMMU | 66.1% | 72.7% | GPT-4.1 mini+6.6 |
| Instruction followingNeither model reports this | — | — | Not comparable |
| Human preferenceNeither model reports this | — | — | Not comparable |
| Categories wonOut of 2 comparable | 0 | 2 | GPT-4.1 mini |
Each category averages only the benchmarks every model here reports, so no one is credited for a test the other did not run. A category with no shared test is marked Not comparable rather than guessed at.
Side by side
Mistral Medium 3 and GPT-4.1 mini, row by row
| Attribute | Mistral Medium 3Mistral AI | GPT-4.1 miniOpenAI |
|---|---|---|
| Specification | ||
| MakerWho built it | Mistral AI | OpenAI |
| Released | 2025-05 | 2025-04 |
| Context windowHow much can go in at once | 131,072 tokens | 1,047,576 tokens |
| Max output | 8,192 tokens | 32,768 tokens |
| Input | Text, Image | Text, Image |
| ReasoningSpends extra tokens thinking before it answers | No | No |
| Tool calling | Yes | Yes |
| Knowledge cutoff | Not reported | 2024-06 |
| Licence | Proprietary | Proprietary |
| Open weightsCan you download and run it yourself | No | No |
| Price | ||
| Input priceUSD per million tokens in | $0.40 | $0.40 |
| Output priceUSD per million tokens out | $2.00 | $1.60 |
| Cached input | Not reported | $0.10 |
| Blended 3:1A 3-in-to-1-out million-token mix — a fairer single number than input price alone | $0.80 | $0.70Cheapest |
| Published benchmarks | ||
| MMLU-Pro12,000 reasoning-heavy multiple-choice questions across 14 academic subjects, with ten options instead of four. The harder successor to MMLU. | Not reported | |
| GPQA Diamond198 graduate-level physics, chemistry and biology questions written to be Google-proof. PhD holders in the matching field score about 65%. | ||
| SWE-bench Verified500 human-validated GitHub issues from real Python repositories. The model must produce a patch that makes the project's own tests pass. | Not reported | |
| HumanEval164 short Python functions written from a docstring. Saturated at the frontier — kept here for continuity with older models. | Not reported | |
| MMMUCollege-level questions that require reading charts, diagrams, tables and photographs alongside the text. | ||
| Links | Full page | Full page |
| Row verified | 2026-08 | 2026-08 |
Where these numbers come from
Every score on this page is a published figure, taken from the model's own card, system card, technical report or release post, or from a public leaderboard. CorX Labs did not run these evaluations. Most are self-reported by the lab that built the model, which means they were produced under that lab's own choice of prompt, scaffold and number of attempts — so treat them as a starting point for a shortlist, not as a settled ranking.
A score someone other than the model's maker measured is marked Independent and names its measurer. Those are the stronger numbers on this page — an outside harness has no reason to flatter anyone — and there are not many of them.
Where a figure has not been published, the cell reads Not reported rather than an estimate. Nothing here is inferred, interpolated or guessed. Each model records the month its row was last checked. Full method and caveats.
Questions
Mistral Medium 3 or GPT-4.1 mini?
Is Mistral Medium 3 better than GPT-4.1 mini?
GPT-4.1 mini wins 2 of the 2 benchmarks both models report, Mistral Medium 3 wins 0, consistently. The average gap across shared tests is 4.6 points.
Which is cheaper, Mistral Medium 3 or GPT-4.1 mini?
On a 3:1 input-to-output mix, GPT-4.1 mini costs $0.70 per million tokens against $0.80 for Mistral Medium 3 — a small difference.
What is the difference between Mistral Medium 3 and GPT-4.1 mini?
GPT-4.1 mini takes 1M tokens of context against 131K — 8.0× more room for long documents or a large codebase.