Project AURIONAI Use-Cases Research & Innovation
Mistral AI
multimodal LLM

Mistral Small 4

Mistral model that unifies instruct, reasoning and coding.

Mistral AIMistral AIFrance
Internal benchmark.
Levelmodel
Classmultimodal LLM
Parameters119B
Context256,000
ReleasedMar 2026
Opennessopen weights permissive
Runs onmulti-GPU node, single datacentre GPU
Statusactive

From the model's own page; who made each picture and where it comes from are listed on the Image Credits page.

Benchmark Results

vendor-reportedmaintainer-run
GPQA Diamondknowledge and reasoning
vendor-reported
71.2
050100

Score in %. Higher is better.

huggingface.co
SEA-HELMSoutheast Asian and Indonesian
Checkpoint Mistral-Small-4-119B-2603 as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstrapsmaintainer-run
63.14 95% CI [-1.42, +1.50]
04080

Score in score. Higher is better.

SEA-HELM leaderboard, 58 models, Indonesian (ID)18 Sep 2026 leaderboard.sea-lion.ai
Checkpoint Mistral-Small-4-119B-2603 as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstrapsmaintainer-run
52.07 95% CI [-0.61, +0.64]
04080

Score in score. Higher is better.

SEA-HELM leaderboard, 58 models, SEA average18 Sep 2026 leaderboard.sea-lion.ai
Where this comes from
Every score names who ran it: vendor-reported, maintainer-run or independent. A rank is shown with the population it was ranked in (gap rule 3).

Lineage

finetune
Mistral Small 4 to Leanstral-1.5-119B-A6B: finetune, medium confidenceMistral Small 4Mistral Small 4Mistral AI, Mar 2026Leanstral-1.5-119B-A6BLeanstral-1.5-119B-A6BMistral AI
Mistral AILeanstral-1.5-119B-A6B is a child (finetune)medium confidencenotes
Where this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. The tree shows two generations each way, 2 models. Confidence and evidence are the notes' own.

R&D Topics

Open-weight frontier releases as research substrateused byReasoning via large-scale RLproduced by

Identity

Developer
Mistral AI
Family
not in the notes
Released
Mar 2026 sourcesnotes
Weights published
–
Record checked
10 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
MoE MoE (128 experts, 4 active)
Total parameters
119B sources
Active parameters
6.5B
Checkpoint parameters
–
Context length
256,000 tokens
Input
text, image
Output
text
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
open weights permissive
Licence
Apache License 2.0 sources
Access routes
open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
multi-GPU node, single datacentre GPU
Footprint
T3 estimate (unverified).
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
not stated
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
Indonesian not named on card
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Industry Use Cases

No industry use case links this model.

Sources

  1. Models Overview docs.mistral.ai https://docs.mistral.ai/getting-started/models/models_overview/
  2. huggingface.co/mistralai https://huggingface.co/mistralai
  3. huggingface.co/mistralai/Mistral-Small-4-119B-2603 https://huggingface.co/mistralai/Mistral-Small-4-119B-2603

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Mistral Small 4", https://aurion.semai.id/models/mistral-ai-mistral-small-4/, records of October 2026.