ZMIME
Mistral AI · Released Oct 16, 2024

Ministral 3B

Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads.

  • Open weights
  • Tool calling
Capability
118.1ECI · #144 of 148
Input
$0.10per 1M tokens
Output
$0.10per 1M tokens
Context
128K8K max output
Pricing

What it costs

Input
$0.10per million tokens
Output
$0.10per million tokens
Blended (3:1)
$0.10cheaper than 91% of priced models

Typical monthly bills

Monthly usageEstimated cost
Side project2M input + 0.5M output tokens$0.25
Team assistant25M input + 5M output tokens$3.00
Production app250M input + 50M output tokens$30.00

Median of 1 paid API provider on models.dev.

Capability

Independent benchmarks

118.1Epoch Capabilities Index
#144 of 148 scored models
88Median 146167

The shaded band is Epoch AI’s confidence range (107.4–121.8); the tick marks the median scored model.

  • GPQA DiamondGraduate-level science questions25.3%

Scores from Epoch AI, run independently of Mistral AI.

Specifications

The details

Lab
Mistral AI
Released
Oct 16, 2024
Knowledge cutoff
Mar 2024
Context window
128,000 tokens
Max output
8,192 tokens
Inputs
Text
Output
Text
Reasoning
No
Tool calling
Yes
Structured output
No
Weights
Open weights
Availability
1 API providerlisted on models.dev
Same lab

More from Mistral AI

Questions

About Ministral 3B

How much does Ministral 3B cost?

Ministral 3B costs $0.10 input / $0.10 output per million tokens (median across 1 API provider). At a 3:1 input-to-output mix that is $0.10 per million tokens, cheaper than 91% of the 360 priced models we track.

What is the context window of Ministral 3B?

Ministral 3B accepts up to 128,000 tokens per request and can write up to 8,192 tokens in one response.

How good is Ministral 3B?

Epoch AI gives Ministral 3B a Capabilities Index score of 118.1 (likely range 107.4–121.8), ranking it #144 of 148 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 25.3%.

Is Ministral 3B open source?

Yes. Mistral AI publishes the weights, so it can be downloaded and self-hosted.

What inputs does Ministral 3B support?

Ministral 3B accepts text and replies in text. It is not a dedicated reasoning model, supports tool calling.

When was Ministral 3B released?

Mistral AI released Ministral 3B on Oct 16, 2024. Its training data runs to Mar 2024.

What are the best alternatives to Ministral 3B?

The closest current models from other labs on capability, price and release date are Llama-3.1-8B-Instruct (Meta, ECI 116.6, $0.152 / $0.167), Qwen2.5 7B Instruct (Alibaba (Qwen), ECI 118.5, $0.175 / $0.70), Gemma 3 4B IT (Google, ECI 116.1, $0.05 / $0.10) and Claude Haiku 3 (Anthropic, ECI 118.4, $0.25 / $1.25).