Mistral Large 3
Mistral · 1 configuration
Intelligence9.3
Input / 1M$0.50
Output / 1M$1.50
Context256K
Tariff: Mistral · Source: artificial-analysis · USD per million tokens
Thinking configuration
What does more thinking buy you?
Compare this model’s measured thinking configurations.
Scroll the chart ↔ · tap a point for details
Efficient frontier: best measured score at each cost1 measured configurations · 0 without both values
Selected: thinking default · Click a point to switch configuration
Flagship Mistral model for advanced reasoning, coding, and multilingual work
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
9.3 pointsGPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
68.0 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
4.2 %SciCodeCode generation for scientific research problems.
36.6 %Terminal-Bench 2.1Agent tasks in a terminal environment. Results also depend on the evaluation harness.
12.0 %τ²-BenchTool use and interaction in assistance scenarios. Does not replace an evaluation of your integration.
24.6 %IFBenchVerifiable instruction following. A proxy for precision rather than a benchmark of creative style.
36.2 %MMMU-ProMultimodal understanding of academic problems that include images.
55.7 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
24.9 %Specifications from the catalog
Inputtext, image
Outputtext
Tool callingNo
Structured outputNot available
Speed78.4 tok/s
First token1.04 s
Maximum output256K
Catalog listing date2025-12-02