mistral-large-3

apache-2.0675B

A general-purpose multimodal mixture-of-experts model for production-grade tasks and enterprise workloads.

Context
262K

Capabilities

Infrastructure
Streaming
Fine-tuning

Details

Provider Ollama
Creator Mistral AI
Familymistral-large
Licenseapache-2.0
Status—
Typechat
Input modalitiestext
Output modalitiestext
Open weightYes
Parameters675B
Architecture—
Context window262K tokens
Max input—
Max output—
Reasoning tokens—
Knowledge cutoff—
Release date
Sourceofficial

Family Comparison: mistral-large

ModelContextMax out
mistral-large-3262K—
mistral-large-41M—

API

GET/v1/models/ollama/mistral-large-3

mistral-large-3

A general-purpose multimodal mixture-of-experts model for production-grade tasks and enterprise workloads.

Changes · 4 entries
mistral-large-3update1cd659eMar 24, 2026, 04:19 AM
capabilitiesfine tuningyes
mistral-large-3update2b100acMar 23, 2026, 05:30 AM
descriptionA general-purpose multimodal mixture-of-experts model for...
release_date2025-12-02
parameters675
page_urlhttps://ollama.com/library/mistral-large-3
taglineA general-purpose multimodal mixture-of-experts model for...
context_window262144
licenseapache-2.0
open_weightyes
modalitiesinput["text"]output["text"]
mistral-large-3update7a4f1f6Mar 22, 2026, 05:43 AM
model_typechat
mistral-large-3create7f5c66fMar 21, 2026, 05:16 AM