Mistral Large 4
Made by Mistral AI, released 6 Oct 2026. Epoch AI does not state how it can be accessed. No Top tier or Strong results yet among the models listed here.
In short
What it does well
No Top tier or Strong ratings on the four main tests yet.
Where it falls short
- Knowledge: Behind the leaders (42nd of 43 listed models)
- Writing: Behind the leaders (74th of 77 listed models)
- Coding: Middle (45th of 77 listed models)
- Maths: Middle (31st of 43 listed models)
What we don't know yet
- Fixing real software bugs: Not enough data
The four ratings
- Top tier
- Strong
- Middle
- Behind the leaders
- Not enough data
Ratings compare only the 77 models listed here, in quarters, best first. “Not enough data” means no result yet. It is never a low score.
- Writing
- Behind the leadersHow often people preferred its answers to writing prompts. 1,361 on Arena's scale · Source for Mistral Large 4 writing score (lmarena)
- Coding
- MiddleHow often people preferred its answers to coding questions. 1,478 on Arena's scale · Source for Mistral Large 4 coding score (lmarena)
- Maths
- MiddleHow many very hard maths problems it solved. 58.6% · Source for Mistral Large 4 maths score (epoch)
- Knowledge
- Behind the leadersHow many short factual questions it answered correctly. 20.0% · Source for Mistral Large 4 knowledge score (epoch)
Evidence completeness: Broad evidence, 12 of 13 tests have a result.
Where it is relatively stronger or weaker
This compares each kind of request with the model's own overall position among the 77 listed models, using Arena's votes. It is about performance on different languages, kinds of request and topics, not about political or social bias.
Relatively stronger
- Spanish: 21st of 70 on questions asked in Spanish, against 68th of 77 overall. Source for Spanish
- French: 22nd of 72 on questions asked in French, against 68th of 77 overall. Source for French
- Chinese: 49th of 76 on questions asked in Chinese, against 68th of 77 overall. Source for Chinese
Relatively weaker
No category stands out as clearly weaker than its overall position.
Show every category we can see
| Category | Position | Arena score | Votes | Source |
|---|---|---|---|---|
| Expert-level prompts Kind of request | 63rd of 77 | 1,447 95% range 1,430 to 1,464 | 1,249 | Source for Expert-level prompts |
| Hard prompts Kind of request | 66th of 77 | 1,444 95% range 1,436 to 1,451 | 7,553 | Source for Hard prompts |
| Long prompts Kind of request | 66th of 77 | 1,428 95% range 1,419 to 1,437 | 5,175 | Source for Long prompts |
| Longer conversations Kind of request | 65th of 77 | 1,437 95% range 1,423 to 1,452 | 1,773 | Source for Longer conversations |
| Chinese Language | 49th of 76 | 1,492 95% range 1,473 to 1,510 | 1,055 | Source for Chinese |
| English prompts Language | 69th of 77 | 1,432 95% range 1,423 to 1,442 | 4,713 | Source for English prompts |
| French Language | 22nd of 72 | 1,489 95% range 1,460 to 1,517 | 483 | Source for French |
| Non-English prompts Language | 66th of 77 | 1,420 95% range 1,413 to 1,428 | 7,344 | Source for Non-English prompts |
| Russian Language | 69th of 77 | 1,414 95% range 1,400 to 1,429 | 1,653 | Source for Russian |
| Spanish Language | 21st of 70 | 1,470 95% range 1,442 to 1,499 | 423 | Source for Spanish |
| Entertainment, sport and media Topic | 71st of 77 | 1,376 95% range 1,365 to 1,388 | 3,181 | Source for Entertainment, sport and media |
| Law and government Topic | 68th of 76 | 1,428 95% range 1,408 to 1,447 | 974 | Source for Law and government |
| Medicine and health Topic | 64th of 75 | 1,428 95% range 1,408 to 1,448 | 865 | Source for Medicine and health |
| Science Topic | 65th of 77 | 1,443 95% range 1,429 to 1,456 | 1,958 | Source for Science |
| Software and IT Topic | 52nd of 77 | 1,467 95% range 1,458 to 1,477 | 4,177 | Source for Software and IT |
Bias: what we can and can't see
Our sources do not measure political lean, stereotyping, flattery (agreeing with you too readily) or safety behaviour, so we cannot say how any model does on those. What we can show is above: where a model is relatively stronger or weaker across languages, kinds of request and topics. Also, people voting in Arena can favour answer style, such as length and formatting, and our numbers do not adjust for that. See what AI bias means and what these tests can tell you.
Best-supported use cases
This model is not among the options we recommend for any of the five tasks. That reflects the rules and the models listed, not a verdict on the model.
Where an alternative might be more suitable
- Writing and creating content: rated Behind the leaders on writing votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Coding and building things: rated Middle on coding votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Homework and maths: rated Middle on hard maths problems. Compare it with GPT-6.1 Sol, the best-supported choice for this task.
- Business emails and everyday work: rated Middle on business-task votes. Compare it with Gemini 4 Argon, the best-supported choice for this task.
- Research and checking facts: rated Behind the leaders on short factual questions. Compare it with Gemini 4 Argon, the best-supported choice for this task.
What it is
- Maker
- Mistral AI
- Release date (Epoch AI)
- 6 Oct 2026
- Access, as listed by Epoch AI
- Not stated
- Licence, as listed by Arena
- Proprietary
- Maker's country (Epoch AI)
- France
- Last observation in our data
- 28 Oct 2026
- Data last changed
- 29 Oct 2026, 14:36 UTC
Arena lists 'Mistral Large 4'; Epoch's id is 'mistral-large-4-0'. Matched by name and release family.
We have not verified an official product page for this model, so we do not link to one.
How you can use it
Proprietary. The model's files are not published, so it is generally available only through its maker or partners the maker chooses.
How you can reach it (an app, a website, a developer API or a third-party tool) and what that costs are not covered by our benchmark data. Where we could verify prices from the maker's own pages, they are shown under Price. See models, apps and harnesses and how AI pricing works.
Price
Price not independently verified
See Mistral plans and pricing (checked 28 Oct 2026). Page opened and read. It does not name the models each plan includes.
We never infer subscription prices from API prices. New to “tokens” and “per million”? See how AI pricing works.
Good to remember
- Benchmarks are samples of tasks. None proves performance in every situation.
- Arena scores come from people choosing between answers. That measures preference, not correctness.
- Ratings are relative to the 77 models listed here. A different set would change them.
- Sources test models at different settings. Each result shows the setting used, and we do not adjust it.
The evidence, for the curious
Show all 13 tests with settings and sources
| Test | Result | Rating | Setting used | Source identifier | Observed | Links |
|---|---|---|---|---|---|---|
| Arena text: overall preference Arena 'text' leaderboard (Bradley-Terry rating, no style control) |
1,430 Arena score 95% range 1,424 to 1,436, 12,066 votes |
Behind the leaders | effort: not stated, bradley-terry rating | Mistral Large 4 lmarena · text/latest |
28 Oct 2026 | Source for Mistral Large 4 Arena text: overall preference score (lmarena) |
| Arena text: creative writing Arena 'text' leaderboard, category creative_writing |
1,361 Arena score 95% range 1,349 to 1,373, 2,736 votes |
Behind the leaders | effort: not stated, bradley-terry rating | Mistral Large 4 lmarena · text/latest |
28 Oct 2026 | Source for Mistral Large 4 Arena text: creative writing score (lmarena) |
| Arena text: coding Arena 'text' leaderboard, category coding |
1,478 Arena score 95% range 1,467 to 1,490, 2,770 votes |
Middle | effort: not stated, bradley-terry rating | Mistral Large 4 lmarena · text/latest |
28 Oct 2026 | Source for Mistral Large 4 Arena text: coding score (lmarena) |
| Arena text: maths Arena 'text' leaderboard, category math |
1,468 Arena score 95% range 1,445 to 1,490, 645 votes |
Middle | effort: not stated, bradley-terry rating | Mistral Large 4 lmarena · text/latest |
28 Oct 2026 | Source for Mistral Large 4 Arena text: maths score (lmarena) |
| Arena text: following instructions Arena 'text' leaderboard, category instruction_following |
1,419 Arena score 95% range 1,409 to 1,428, 4,230 votes |
Behind the leaders | effort: not stated, bradley-terry rating | Mistral Large 4 lmarena · text/latest |
28 Oct 2026 | Source for Mistral Large 4 Arena text: following instructions score (lmarena) |
| Arena text: business and management Arena 'text' leaderboard, category industry_business_and_management_and_financial_operations |
1,439 Arena score 95% range 1,425 to 1,452, 2,060 votes |
Middle | effort: not stated, bradley-terry rating | Mistral Large 4 lmarena · text/latest |
28 Oct 2026 | Source for Mistral Large 4 Arena text: business and management score (lmarena) |
| Arena WebDev: building web apps Arena 'webdev' leaderboard (Code Arena), category overall |
1,544 Arena score 95% range 1,534 to 1,555, 4,068 votes |
Strong | effort: not stated, bradley-terry rating | Mistral Large 4 lmarena · webdev/latest |
28 Oct 2026 | Source for Mistral Large 4 Arena WebDev: building web apps score (lmarena) |
| FrontierMath Tiers 1-3 (v2) FrontierMath Tiers 1-3 v2 (private set), benchmark released 2026-06-12 |
58.6% share of problems solved standard error ±2.9 points |
Middle | effort: high, best score across scorers | mistral-large-4-0_high epoch · frontiermath_tiers_1_3_v2.csv |
12 Oct 2026 | Source for Mistral Large 4 FrontierMath Tiers 1-3 (v2) score (epoch) No public run log in the data file. |
| OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025, benchmark released 2024-12-19 |
98.3% share of problems solved standard error ±0.6 points |
Strong | effort: high, best score across scorers | mistral-large-4-0_high epoch · otis_mock_aime_2024_2025.csv |
12 Oct 2026 | Source for Mistral Large 4 OTIS Mock AIME 2024-2025 score (epoch) No public run log in the data file. |
| GPQA Diamond GPQA Diamond, benchmark released 2023-11-20 |
88.1% share of questions correct standard error ±1.9 points |
Middle | effort: high, best score across scorers | mistral-large-4-0_high epoch · gpqa_diamond.csv |
12 Oct 2026 | Source for Mistral Large 4 GPQA Diamond score (epoch) No public run log in the data file. |
| SimpleQA Verified SimpleQA Verified, benchmark released 2025-09-09 |
20.0% share of questions correct standard error ±1.3 points |
Behind the leaders | effort: high, best score across scorers | mistral-large-4-0_high epoch · simpleqa_verified.csv |
9 Oct 2026 | Source for Mistral Large 4 SimpleQA Verified score (epoch) No public run log in the data file. |
| SWE-bench Verified (Epoch runs) | Not enough data | |||||
| FrontierCode (reported by Cognition) FrontierCode, benchmark released 2026-06-08, aggregation Mean@5 |
39.4% mean score over 5 attempts no uncertainty figure published |
Not rated (results use different tools) | effort: high, tool: chisel, mean@5 | mistral-large-4-0_high epoch · frontiercode_external.csv |
not provided | Source for Mistral Large 4 FrontierCode (reported by Cognition) score (epoch) · Run details for FrontierCode (reported by Cognition) |
Missing evidence
Not enough data for these tests:
- SWE-bench Verified (Epoch runs)
Missing results are not low scores and do not count against the model.
Where a source gives no public run log, we cannot show the run behind the score. See data sources.
Not sure this is the right one?
Tell us what you need help with and we'll narrow the options down.