MMODELYST
Models/Magistral Small 1.2
MDL

Magistral Small 1.2

Mistral

Verified against Artificial Analysis · Jul 13, 2026
Capability
61
percentile index
Price /M
$0.5 / $1.5
in / out
Context
Avg score
43.0
10 benchmarks
AA Intelligence
11.3
index
Speed
84 tok/s
TTFT 0.4s
Released
Sep 17, 2025
About

Magistral Small 1.2 is a proprietary model in the Magistral family from Mistral. Benchmarked on 10 evals, averaging 43.0.

Benchmark scores · 10
avg 43.0 — higher is better
AIME
80.3
MMLU-Pro
76.8
LiveCodeBench
72.3
GPQA Diamond
66.3
IFBench
44.4
SciCode
35.2
τ²-bench
27.8
AA Long-Context Reasoning
16.3
Humanity's Last Exam
6.1
Terminal-Bench Hard
4.5
Details
PublisherMistral
FamilyMagistral
Parameters
License
WeightsClosed
ReleasedSep 17, 2025
Performance & price
Output speed84 tok/s
Latency (TTFT)0.4s
First answer24s
Price in / out$0.5 / $1.5 per 1M
Blended price$0.75 / 1M
Coding index14.7
Math index80.3

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$1.5/M