MMODELYST
Models/Mistral Small 3.1
MDL

Mistral Small 3.1

Mistral

Verified against Artificial Analysis · Jul 13, 2026
Capability
42
percentile index
Price /M
$0.1 / $0.3
in / out
Context
Avg score
29.1
11 benchmarks
AA Intelligence
14.7
index
Speed
160 tok/s
TTFT 0.5s
Released
Mar 17, 2025
About

Mistral Small 3.1 is a open-weights model in the Mistral family from Mistral. Benchmarked on 11 evals, averaging 29.1.

Benchmark scores · 11
avg 29.1 — higher is better
MATH
70.7
MMLU-Pro
65.9
GPQA Diamond
45.4
IFBench
29.9
SciCode
26.5
τ²-bench
25.1
LiveCodeBench
21.2
AA Long-Context Reasoning
19.7
Terminal-Bench Hard
7.6
Humanity's Last Exam
4.8
AIME
3.7
Details
PublisherMistral
FamilyMistral
Parameters
License
WeightsOpen
ReleasedMar 17, 2025
Performance & price
Output speed160 tok/s
Latency (TTFT)0.5s
Price in / out$0.1 / $0.3 per 1M
Blended price$0.15 / 1M
Coding index26.3
Math index3.7

Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →

Capability over time

Append-only ledger — every observed change to this model's numbers.

API providers
Artificial Analysis$0.3/M