Mistral AI · Ministral 3 3b Base 2512 · active

Ministral-3-3B-Base-2512

Ministral-3-3B-Base-2512 is an open-weight vision-language model published by Mistral AI.

mistralai/Ministral-3-3B-Base-2512
Statusactive
InputText, Image
OutputText
Context256K
Total / active parameters4.3B /
Licenseapache-2.0
APIUnknown
Open weights / self-hostYes / Yes

This page represents one immutable developer model ID. Serving endpoints, pricing, regions, quantizations, and service tiers attach separately and never create duplicate model records.

Provider availability

Serving endpoints and pricing

No directly sourced provider offerings yetPricing and availability stay unknown until collected from each serving provider.

Technical specification

Developer model IDmistralai/Ministral-3-3B-Base-2512
VersionMinistral-3-3B-Base-2512
ReleasedNot reported
ArchitectureMistral3ForConditionalGeneration
Maximum outputUnknown
Knowledge cutoffNot reported

Capabilities

generation
ReasoningNo
VisionYes
Tool callingNo

Version lineage

PredecessorUnknown
SuccessorsUnknown
Aliasesmistralai/Ministral-3-3B-Base-2512, mistralai/Ministral-3-3B-Base-2512

Explicit unknowns

  • Release date
  • Maximum output
  • Knowledge cutoff
  • API availability
  • Provider pricing

Comparable peers

Related model comparisons

All comparisons →

Questions

Ministral-3-3B-Base-2512 FAQs

What is Ministral-3-3B-Base-2512?+

Ministral-3-3B-Base-2512 is an open-weight vision-language model published by Mistral AI. It is developed by Mistral AI and its current sourced lifecycle status is active.

What inputs and outputs does Ministral-3-3B-Base-2512 support?+

Ministral-3-3B-Base-2512's current record lists Text and Image as input and Text as output.

How much context does Ministral-3-3B-Base-2512 support?+

The current record lists a 262,144-token context window and does not report a maximum output length.

Is Ministral-3-3B-Base-2512 available through an API?+

Ministral-3-3B-Base-2512's current primary-source record does not establish API availability. No provider endpoint is treated as available without direct provider evidence.

How large is Ministral-3-3B-Base-2512?+

The current record lists 4.3B total parameters and an unknown active parameter count. Its architecture is Mistral3ForConditionalGeneration.

What capabilities and tasks does Ministral-3-3B-Base-2512 support?+

Ministral-3-3B-Base-2512's recorded capabilities are generation. Its supported tasks are image and text.

Can Ministral-3-3B-Base-2512 be self-hosted?+

Ministral-3-3B-Base-2512 is an open-weight model. Self-hosting is supported, and the recorded license is apache-2.0.

Does Ministral-3-3B-Base-2512 have related versions or aliases?+

No predecessor is recorded. No successor is recorded. Recorded aliases are mistralai/Ministral-3-3B-Base-2512.

Which models can Ministral-3-3B-Base-2512 be compared with?+

The current compatible peer set includes Ministral-3-3B-Reasoning-2512, Ministral-3-3B-Instruct-2512, Ministral-3-8B-Base-2512, granite-vision-4.1-4b, and Phi-4-multimodal-instruct. Compatibility requires shared sourced modalities and tasks; it does not imply that one model is better. Browse model comparisons

How current is the Model Markets record for Ministral-3-3B-Base-2512?+

Ministral-3-3B-Base-2512 was last verified Aug 28, 2026 from Mistral AI's primary source, “mistralai/Ministral-3-3B-Base-2512 developer model repository.” The evidence is classified as official fact.

Primary evidence

Sources and observation date