MistralReleased on Dec 4, 2025

Ministral 3 (14B Reasoning 2512): Benchmarks, Pricing & Context Window

Ministral 3 (14B Reasoning 2512) is a language model from Mistral, released in December 2025, with multimodal input.

A balanced model in the Ministral 3 family, Ministral 3 14B is a powerful, efficient tiny language model with vision capabilities. This model is the reasoning post-trained version, trained for reasoning tasks, making it ideal for math,

Input
TextImage
Output
Text

Ministral 3 (14B Reasoning 2512) pricing

Providers

Ministral 3 (14B Reasoning 2512) starts at $0.200 per million input tokens and $0.200 per million output tokens via Mistral AI.

ProviderInput $/MOutput $/MMax InputMax OutputLatency sThroughputQuantInputOutput
Mistral AI logoMistral AI
$0.200$0.200262.1K262.1K
0.23
129 c/s

Ministral 3 (14B Reasoning 2512) API

API access coming soon

Ministral 3 (14B Reasoning 2512) will be available through our gateway shortly.

Ministral 3 (14B Reasoning 2512) examples

Recent arena outputs from Ministral 3 (14B Reasoning 2512), picked from the highest-ranked matchups.

Ministral 3 (14B Reasoning 2512) license

Ministral 3 (14B Reasoning 2512) is released under the Apache 2.0 license, which permits commercial use, has 14.0B parameters.

License
Apache 2.0
Commercial use allowed
Parameters
14.0B

Apache License 2.0 - allows commercial use

FAQ

Common questions about Ministral 3 (14B Reasoning 2512).

What is the Ministral 3 (14B Reasoning 2512) release date?

Ministral 3 (14B Reasoning 2512) was released on December 4, 2025 by Mistral.

Who created Ministral 3 (14B Reasoning 2512)?

Ministral 3 (14B Reasoning 2512) was created by Mistral.

How many parameters does Ministral 3 (14B Reasoning 2512) have?

Ministral 3 (14B Reasoning 2512) has 14.0 billion parameters.

What is the license for Ministral 3 (14B Reasoning 2512)?

Ministral 3 (14B Reasoning 2512) is released under the Apache 2.0 license. This is an open-source/open-weight license.

Is Ministral 3 (14B Reasoning 2512) multimodal?

Yes, Ministral 3 (14B Reasoning 2512) is a multimodal model that can process both text and images as input.