Mistral Large 4 vs Open Mistral 7B
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models
Mistral Large 435
Open Mistral 7B—
Terminal-Bench 4.0 Terminal-Bench 4.0 · source setting 2: Agents use a terminal to complete executable coding, system administration and research tasks in the 4.0 suite
Mistral Large 426.8
Open Mistral 7B—
AutomationBench-AA AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation
Mistral Large 459.9
Open Mistral 7B—
Arena Text · creative writing Arena Text · creative writing: Human pairwise preferences in the stated text vote pool
Mistral Large 41,368
Open Mistral 7B—
MMMU-Pro (vision) MMMU-Pro (vision): College-level questions that require understanding images: charts, diagrams, photos
Mistral Large 476.4
Open Mistral 7B—
| Benchmark | Mistral Large 4 | Open Mistral 7B |
|---|---|---|
| Humanity's Last Exam | 35 | — |
| Terminal-Bench 4.0 | 26.8 | — |
| AutomationBench-AA | 59.9 | — |
| Arena Text · creative writing | 1,368 | — |
| MMMU-Pro (vision) | 76.4 | — |
Specs
| Spec | Mistral Large 4 | Open Mistral 7B |
|---|---|---|
| Input, per 1M | $1.36 | — |
| Output, per 1M | $4.18 | — |
| Cost per task? | $1.13 | — |
| Speed, tok/s | 67 | — |
| Context | 1M | — |
| Weights | Closed | Not verified |
| Released | Oct 2026 | — |
| See images | ✓ | ? |
| Read PDFs | ✓ | ? |
| Hear audio | — | ? |
| Watch video | — | ? |
| Call tools | ✓ | ? |
| Force a tool call | Partial | ? |
| Return strict JSON | ✓ | ? |
| Search the web | ? | ? |
| Run code | ? | ? |
| Use a computer | ? | ? |
What breaks when you switch
Mistral Large 4 → Open Mistral 7B
Context: not verified
Open Mistral 7B → Mistral Large 4
Infinite whitespace stream›
In JSON Object mode, include JSON in the system or user prompt to avoid a possible infinite whitespace stream.
Content is a list of chunks›
With reasoning_effort high, message.content is a chunk list rather than a plain string; parse thinking and text chunks.
Only function calling · Mistral regional endpoints›
Regional endpoints support function calling only and omit Agents, Batch and Files; check regional model availability before routing requests.
10% upcharge · Mistral regional endpoints›
Regional inference charges 1.1× standard list prices for input, output, cached reads and cache writes.
Reinforcement learning run still in flight›
The public preview reinforcement learning run is still in flight, and Mistral expects further rapid changes as training scales.