Mistral Small 3

Name: Mistral Small 3 benchmark results
Creator: BenchLeader

Best configuration ranks #553 of 610 on the BenchLeader Index at 38.2 ±4.5. Last measured 30 Jan 2025stale: no new result in six months. Released 30 Jan 2025.

Blended price: $0.150/M; $0.100 in · $0.300 out
Output speed: 152 tok/s
First answer: 0.76 s; first token 0.76 s
Context: 32k

How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.

Overall index38
Reasoning36
Maths27
Instruction following31
Long context25
Composite37

Benchmark results

One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

Benchmark	default	Source
GPQA Diamond	47.3%#212	Epoch AI Benchmarking Hub
GPQA Diamond (AA)not in index	46.2%#425	Artificial Analysis
Humanity's Last Exam (AA)not in index	3.8%#483	Artificial Analysis

Agents & tools

Benchmark	default	Source
τ²-Bench Telecom (AA)not in index	19.6%#323	Artificial Analysis

Maths

Benchmark	default	Source
OTIS Mock AIME	6.7%#220	Epoch AI Benchmarking Hub

Instruction following

Benchmark	default	Source
IFBench	26.4%#367	Artificial Analysis

Long context

Benchmark	default	Source
AA-LCR	0.0%#419	Artificial Analysis

Composite

Benchmark	default	Source
Epoch Capabilities Indexnot in index	127.1#145	Epoch AI Benchmarking Hub
AA Intelligence Index	6.7#431	Artificial Analysis

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.

Workload	Tokens in / out	Cost	With caching	Time
Chat reply	400 / 300	$0.0001	–	2.7 s
Summarise a 30-page report	12,000 / 600	$0.0014	–	4.7 s
Code edit	6,000 / 1,500	$0.0011	–	10.6 s
Agentic coding session	60,000 / 4,000	$0.0072	–	27.0 s
Structured extraction	2,000 / 200	$0.0003	–	2.1 s