Mistral AIFrance· released Dec 2025· open weights

Mistral Large 3

Name: Mistral Large 3
Author: Mistral AI

Provisional: 5 comparable tests so far. Ranked once it reaches 6 across 3 counting categories.

Mistral AI, France. One of 2 models from this lab on the Index.

Third-party publishedLow confidence (thin coverage)

Snapshot September 8, 2026

OpenCharts Index

44.1

Provisional: not enough comparable tests or counting categories to rank yet · 5 comparable tests · 0 counting categories

Drop any single test and the Index lands between 34.3 and 58.8.

Needs more results (0 of 8 so far)

Distance to today's frontier on the hardest tests. Not a measure of intelligence.

What it's known for

Mistral Large 3, in 5 results.

Its widest gap is on CritPt, 100.0% behind the best published result. No comparable results yet in knowledge, math, agentic and long context. Between 50% and 60% behind the best published results, on average.

Strongest category

No category counts yet: a category counts once it has more than one comparable test and the model has taken at least half of them. The scores it has are below.

Best single result

Text Arena

Answer the same prompt as a rival model; real users vote blind on which answer they prefer.

1,414

37% expected win rate against the board leader (1,507) · #77 of 92.

Millions of blind votes on real prompts are the market's measure of what people prefer.

Where it trails

CritPt

Solve an unpublished research-level physics problem to a numeric or symbolic answer checked by an official server.

0.0%

100.0% behind the best published result (32.3%) · #72 of 78.

Research physics is far past textbook recall. Only a few models produce anything a physicist would accept.

Not measured yet

No comparable result yet in these 4 categories. A blank is a blank, never a zero, and it does not lower the Index.

Knowledge
Math
Agentic
Long context

Covered, but not averaged into the Index:

Reasoning1 of 6 tests · needs 3 · mean 0.0
Coding2 of 7 tests · needs 4 · mean 32.9
Multimodalsingle-test category, shown beside the Index · 69.9
Human preferencesingle-test category, shown beside the Index · 73.7

Every score, with its source

See it at work

What Mistral Large 3 was asked to do, and how it did.

Every test it has taken, by category, with the kind of task it faced, who produced the number and where the result landed against the best published one. The examples are original and illustrative, never items from the datasets themselves.

Human preference · mean 73.7 across 1 comparable test · a single-test category, shown beside the Index and never averaged into it

Text ArenaLeaderboard

Answer the same prompt as a rival model; real users vote blind on which answer they prefer.

1,414

37% expected win rate against the board leader (1,507)
#77 of 92

board 2026-09-02

Source: Arena

Every result is read against the best published one on its test: percent scores as a share of the best above chance, Arena ratings as the expected win rate against the board leader, open-ended values on a log scale. Sitting a harder exam never lowers a model. Every number links to the publisher that produced it.

Head to head

Mistral Large 3 against whoever you pick.

Choose a rival. Every counted test both have taken appears side by side as a share of the best published result on that test, with the category means above. A win is a higher share; a gap under 2 points is a tie, because one item on a 45-problem exam is 2.2 points and publishers' own standard errors are of that order.

Mistral Large 3

provisional·Index 44.1D

0 : 3

wins · 0 ties · 3 shared

GPT-6 Astra

#1·Index 99.0A+

GPT-6 Astra comes out ahead on 3 of the 3 tests both have taken.

Category means

Reasoning
0.0
99.6
Knowledge
—
100.0
Coding
32.9
97.7
Math
—
99.8
Agentic
—
98.5
Multimodal
69.9
—
Human preference
73.7
—

Shared tests · biggest gaps first

CritPtReasoning
0.0%
31.7%
−98.2
Code Arena (WebDev)Coding
1,229
1,797
−92.7
SciCodeCoding
36.2%
56.5%
−32.6

Bars are distances to the best published result on each test (percent scores above chance, Arena ratings as win rate against the leader, open-ended values on a log scale). Raw values as the publishers report them.

Open GPT-6 Astra

Scores

Every test, every source

The best published run per benchmark, the raw value as the publisher reports it, who produced the number, how close it comes to the best published result on that test, and where the model sits among every model scored on it. Percent scores are read above chance, Arena ratings as a win rate against the board leader, open-ended values on a log scale. A test with too few models to compare against is shown but not counted.

Category means

Reasoning0.0 · 1/6
Knowledge—
Coding32.9 · 2/7
Math—
Agentic—
Multimodalbeside69.9 · 1/1
Human preferencebeside73.7 · 1/1
Long context—

Mean · tests taken of the category's comparable tests. A faded category either rests on fewer than half of them or is a single-test category (marked “beside”), which is shown next to the Index and never averaged into it.

CritPtLeaderboardReasoning

0.0%

0.0% of the best (32.3%)
0th percentile of 78

Source: Epoch AI — AI Benchmarking Hub

SciCodeLeaderboardCoding

36.2%

58.4% of the best (62.0%)
8th percentile of 73 · n = 338

Source: Epoch AI — AI Benchmarking Hub

Code Arena (WebDev)LeaderboardCoding

1,229

4% expected win rate against the board leader (1,797)
4th percentile of 86

board 2026-09-05