modelgrep
M

Mistral: Mistral Large 3 2512

mistralai/mistral-large-2512

109th smartest of 178ToolsJSONVision
Use via OpenRouter ↗
Intelligence
22.8
109th of 178
Design Elo
1206
Website
Speed
43
193rd fastest
Latency
584ms
first token
Input price
$0.500
163rd cheapest
Context
262K

How it compares

Smarter than39%
of all ranked models
Faster than35%
of all ranked models
Cheaper than45%
of all ranked models

Overview

Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.

Benchmarks

independent · via OpenRouter
Artificial Analysis38th percentile
Intelligence Index
22.8
Coding Index
22.7
Agentic Index
21.7
GPQA Diamond
68%
Humanity's Last Exam
4%
SciCode
36%
Tau²-Bench (agentic)
25%
Design Arena · Elo14,976 tournaments
Website
1206
codecategories
1194
Data Viz
1181
3D
1180
UI Component
1158
Game Dev
1149
asciiart
1111
svg
1049

Providers & pricing (1)

ProviderIn $/MOut $/MUptime
Mistral$0.500$1.5099.3%

Specifications

Context window262K
Max output
Knowledge cutoff
Input modalitiestext, image, file
Output modalitiestext
Prompt caching
Cache read price$0.050/M
ModeratedNo

Mistral Large 3 2512 FAQ

How much does Mistral Large 3 2512 cost?

Mistral Large 3 2512 costs $0.500 per million input tokens and $1.50 per million output tokens via OpenRouter, making it 163rd cheapest of 298 paid models.

How smart is Mistral Large 3 2512?

Mistral Large 3 2512 scores 22.8 on the Artificial Analysis Intelligence Index, ranking 109th of 178 benchmarked models, with a GPQA Diamond score of 68%.

How fast is Mistral Large 3 2512?

Mistral Large 3 2512 generates around 43 tokens per second with 584ms time-to-first-token (p50), the 193rd fastest tracked model.

What is Mistral Large 3 2512's context window?

Mistral Large 3 2512 supports a 262K-token context window. It accepts text, image, file input.

Compare head-to-head