Mistral Small 4
Mistral · 2 thinking configurations
Intelligence11.3
Input / 1M$0.15
Output / 1M$0.60
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 · 1 without both values
Selected: thinking thinking · Click a point to switch configuration
Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system. It combines strong reasoning from...
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
11.3 pointsGPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
76.9 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
9.9 %SciCodeCode generation for scientific research problems.
38.8 %Terminal-Bench 2.1Agent tasks in a terminal environment. Results also depend on the evaluation harness.
21.0 %τ²-BenchTool use and interaction in assistance scenarios. Does not replace an evaluation of your integration.
41.2 %IFBenchVerifiable instruction following. A proxy for precision rather than a benchmark of creative style.
48.2 %MMMU-ProMultimodal understanding of academic problems that include images.
56.8 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
21.7 %Specifications from the catalog
Inputtext, image
Outputtext
Tool callingYes
Structured outputYes
Speed183.5 tok/s
First token0.78 s
Maximum output209.7K
Catalog listing date2026-03-16