Magistral Medium is Mistral AI's language model with a 128K context window and up to 64K output tokens, available from 2 providers, starting at $2.00 / 1M input and $5.00 / 1M output. Mistral AI's mid-tier reasoning model with transparent chain-of-thought capabilities, excelling at multilingual complex problem-solving.
Specifications
Canonical IDmistral-magistral-medium
TypeLanguage
StatusActive
CreatorMistral AIMistral AI
Providers
Context Window128K tokens
Max Output64K tokens
Input ModalitiesImage
Output ModalitiesText
Reasoning Effortsdefault
Release Date · 1 year ago
Benchmarks
Intelligence Index
18.8
#232
Coding Index
16.0
#212
Math Index
40.3
#150
MMLU-Pro
0.8
#159
GPQA
0.7
#203
HLE
0.1
#146
LiveCodeBench
0.5
#130
AIME
0.7
#33
IFBench
0.3
#356
Time to First Token
0.52s
#250
SciCode
0.3
#225
MATH-500
0.9
#57
AIME 2025
0.4
#150
LCR
0.0
#376
TerminalBench Hard
0.1
#199
TAU2
0.2
#271
Output TPS
0.0
#417

Capabilities

Input1/5
Text·
Image
Audio·
Video·
PDF·
Output1/5
Text
Image·
Audio·
Video·
Embedding·
Capabilities4/13
Reasoning
Adaptive Reasoning·
Function Calling
Parallel Function Calling·
Structured Outputs
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill

Pricing by Provider

ProviderStandard
Input
$ / 1M
Output
$ / 1M
Mistral AI logo
Mistral AI
mistral/magistral-medium-latest
$2.00$5.00
Vercel AI Gateway logo
Vercel AI Gateway
mistral/magistral-medium
$2.00$5.00

Cost Calculator

Preset:

Versions

VersionReleasedContextInput / 1MOutput / 1MStatus
Magistral 1.2 Medium40K$2.00$5.00Available
Magistral Medium128K$2.00$5.00Current

Other models

ModelTierReleasedContextInput / 1MOutput / 1M
Magistral 1.2 SmallSmall40K$0.500$1.50
Magistral SmallSmall128K$0.500$1.50

Model IDs