modelgrep
M

Mistral: Ministral 3 8B 2512

mistralai/ministral-8b-2512

145th smartest of 178ToolsJSONVision
Use via OpenRouter ↗
Intelligence
14.8
145th of 178
Design Elo
1112
3D
Speed
11
281st fastest
Latency
268ms
first token
Input price
$0.150
82nd cheapest
Context
262K

How it compares

Smarter than19%
of all ranked models
Faster than5%
of all ranked models
Cheaper than72%
of all ranked models

Overview

A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.

Benchmarks

independent · via OpenRouter
Artificial Analysis19th percentile
Intelligence Index
14.8
Coding Index
10.0
Agentic Index
16.7
GPQA Diamond
47%
Humanity's Last Exam
4%
SciCode
21%
Tau²-Bench (agentic)
27%
Design Arena · Elo1,034 tournaments
3D
1112
Website
1110
codecategories
1105
Game Dev
1060

Providers & pricing (2)

ProviderIn $/MOut $/MUptime
Mistral$0.150$0.15099.6%
NextBitfp8$0.300$0.30099.9%

Specifications

Context window262K
Max output
Knowledge cutoff
Input modalitiestext, image
Output modalitiestext
Prompt caching
Cache read price$0.015/M
ModeratedNo

Ministral 3 8B 2512 FAQ

How much does Ministral 3 8B 2512 cost?

Ministral 3 8B 2512 costs $0.150 per million input tokens and $0.150 per million output tokens via OpenRouter, making it 82nd cheapest of 298 paid models.

How smart is Ministral 3 8B 2512?

Ministral 3 8B 2512 scores 14.8 on the Artificial Analysis Intelligence Index, ranking 145th of 178 benchmarked models, with a GPQA Diamond score of 47%.

How fast is Ministral 3 8B 2512?

Ministral 3 8B 2512 generates around 11 tokens per second with 268ms time-to-first-token (p50), the 281st fastest tracked model.

What is Ministral 3 8B 2512's context window?

Ministral 3 8B 2512 supports a 262K-token context window. It accepts text, image input.

Compare head-to-head