CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.mistral / Ministral-3-3B-Instruct-2512-BF16
From: mistralai <huggingface.co/mistralai>Newsgroups: alt.freeweights.mistralSubject:

Ministral-3-3B-Instruct-2512-BF16

Date: 2026-11-14Commit: b6d637bSigned: 2DC28CC7B26A283B ✓

Model with 3.8B parameters. Safetensors weights, 15.4 GB in all. Published by mistralai under Apache-2.0.

3.8B params15.4GBMinistral-3-3B-Instruct-2512-BF16safetensorspermissive
FactsParams3.8BSize15.4 GBFormatssafetensorsTasknot taggedLicenseApache-2.0From sourcepermissive
$ fw get mistralai/Ministral-3-3B-Instruct-2512-BF16
# manifest sha256 f39d12c52f00… · 15.4 GB · 16 files
$ fw verify ./Ministral-3-3B-Instruct-2512-BF16 mistralai/Ministral-3-3B-Instruct-2512-BF16
From sourceHosted (signed)mistralai/Ministral-3-3B-Instruct-2512-BF16Last check: 2026-11-14 21:24 UTC
Run at homeGet help

Get help: custom fine-tune or private deploy. We review your setup before we quote.

Fine-tune of Ministral-3-3B-Base-2512.

Works with

Run8Fine-tune7Convert8Scan2

Runs with

Ranked by fit to this model (safetensors), then by GitHub stars.

TransformersLocalAItinygradTriton Inference ServerBentoMLCTranslate2ExLlamaV2LitServe
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
15.4 GB
Weights
7.70 GB (safetensors)
Parameters
3.8B
Formats
safetensors
Task
not tagged
License
Apache-2.0 (permissive)
Files
16, each with a sha256
Publisher
mistralai
Added
2026-11-14
Section
alt.freeweights.mistral
Lineage

Original release from the publisher.

Base model: mistralai/Ministral-3-3B-Base-2512

Where these bytes come from
Publisher at the pinned commit→You

Sources are tried in order. A file counts only when its size and sha256 match the manifest. Sources are checked when you ask for them, not on a timer.

Sources
BadgeHostLast checkResult
From sourcehuggingface.co2026-11-14 21:24 UTCpassed

Model card

From the publisher at commit b6d637b. Shown as plain text with links; images open at the publisher.

Ministral 3 3B Instruct 2512 BF16#

The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.

This model is the instruct post-trained version, fine-tuned for instruction tasks, making it ideal for chat and instruction based use cases.

The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware. Ministral 3 3B can even be deployed locally, capable of fitting in 16GB of VRAM in BF16, and less than 8GB of RAM/VRAM when quantized.

We provide a no-loss FP8 version here, you can find other formats and quantizations in the Ministral 3 - Additional Checkpoints collection.

Learn more in our blog post and paper.

Key Features#

Ministral 3 3B consists of two main architectural components:

  • 3.4B Language Model
  • 0.4B Vision Encoder

The Ministral 3 3B Instruct model offers the following capabilities:

  • Vision: Enables the model to analyze images and provide insights based on visual content, in addition to text.
  • Multilingual: Supports dozens of languages, including English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean, Arabic.
  • System Prompt: Maintains strong adherence and support for system prompts.
  • Agentic: Offers best-in-class agentic capabilities with native function calling and JSON outputting.
  • Edge-Optimized: Delivers best-in-class performance at a small scale, deployable anywhere.
  • Apache 2.0 License: Open-source license allowing usage and modification for both commercial and non-commercial purposes.
  • Large Context Window: Supports a 256k context window.
Use Cases#

Ideal for lightweight, real-time applications on edge or low-resource devices, such as:

  • Image captioning
  • Text classification
  • Real-time efficient translation
  • Data extraction
  • Short content generation
  • Fine-tuning and specialization
  • And more...

Bringing advanced AI capabilities to edge and distributed environments for embedded systems.

Ministral 3 Family#

Model NameTypePrecisionLink
Ministral 3 3B Base 2512Base pre-trainedBF16Hugging Face
Ministral 3 3B Instruct 2512Instruct post-trainedBF16Hugging Face
Ministral 3 3B Reasoning 2512Reasoning capableBF16Hugging Face
Ministral 3 8B Base 2512Base pre-trainedBF16Hugging Face
Ministral 3 8B Instruct 2512Instruct post-trainedBF16Hugging Face
Ministral 3 8B Reasoning 2512Reasoning capableBF16Hugging Face
Ministral 3 14B Base 2512Base pre-trainedBF16Hugging Face
Ministral 3 14B Instruct 2512Instruct post-trainedBF16Hugging Face
Ministral 3 14B Reasoning 2512Reasoning capableBF16Hugging Face

Other formats available here.

Benchmark Results#

We compare Ministral 3 to similar sized models.

Reasoning#
ModelAIME25AIME24GPQA DiamondLiveCodeBench
Ministral 3 14B0.8500.8980.7120.646
Qwen3-14B (Thinking)0.7370.8370.6630.593
Ministral 3 8B0.7870.8600.6680.616
Qwen3-VL-8B-Thinking0.7980.8600.6710.580
Ministral 3 3B0.7210.7750.5340.548
Qwen3-VL-4B-Thinking0.6970.7290.6010.513
Instruct#
ModelArena HardWildBenchMATH Maj@1MM MTBench
Ministral 3 14B0.55168.50.9048.49
Qwen3 14B (Non-Thinking)0.42765.10.870NOT MULTIMODAL
Gemma3-12B-Instruct0.43663.20.8546.70
Ministral 3 8B0.50966.80.8768.08
Qwen3-VL-8B-Instruct0.52866.30.9468.00
Ministral 3 3B0.30556.80.8307.83
Qwen3-VL-4B-Instruct0.43856.80.9008.01
Qwen3-VL-2B-Instruct0.16342.20.7866.36
Gemma3-4B-Instruct0.31849.10.7595.23
Base#
ModelMultilingual MMLUMATH CoT 2-ShotAGIEval 5-shotMMLU Redux 5-shotMMLU 5-shotTriviaQA 5-shot
Ministral 3 14B0.7420.6760.6480.8200.7940.749
Qwen3 14B Base0.7540.6200.6610.8370.8040.703
Gemma 3 12B Base0.6900.4870.5870.7660.7450.788
Ministral 3 8B0.7060.6260.5910.7930.7610.681
Qwen 3 8B Base0.7000.5760.5960.7940.7600.639
Ministral 3 3B0.6520.6010.5110.7350.7070.592
Qwen 3 4B Base0.6770.4050.5700.7590.7130.530
Gemma 3 4B Base0.5160.2940.4300.6260.5890.640

License#

This model is licensed under the Apache 2.0 License.

You must not use this model in a manner that infringes, misappropriates, or otherwise violates any third party’s rights, including intellectual property rights.

Files

16 files · 15.4 GBWeights 7.70 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 16 files with their sha256
FileSizesha256
.gitattributes1.62 KB1e11e8e6b9eb852a71eadd8fe191d4b9e329eb81308602dec19ab16da7e172b2
README.md8.57 KB88189c3b964024a8a1d39e0a820596776b997fabb332a151a25973b242f06382
SYSTEM_PROMPT.txt2.41 KB331b249682cd52226c50e533f59825184997f3b31f9a62b2c3fea940db6999c5
chat_template.jinja11.9 KB0701cfbdc2b7d44fdbad104dff604faee4b0543e8247624568777fe465746f9b
config.json1.58 KBc89d1a0b4f237d2892ce911d1fe03e9e5a4834579f7149ebc715a4c3fa564214
consolidated.safetensors7.70 GB722430146632596c6c41137b2e582095ed01a50342932b035cb8662f7855a340
generation_config.json131 Be0923390059f84a9180b00e5501778acc45ea9856cd7f2fd68208b360927c677
model-00001-of-00002.safetensors4.97 GBb3821ebc30884f66e3d26d339e161641f34b91ed916627011e7b08e5f1edd884
model-00002-of-00002.safetensors2.73 GB718d087fa591fd4356b7241f293c24219399d86f092d46cf36f765051498033a
model.safetensors.index.json45.6 KB7829dcf0040e34f1172b401563fcbb27cc3c5a0244ef01e6af18b7a64d63a81e
params.json1.10 KBd15587c57574d2a92c66ff2552c6db5173d4c440ad47f7454dc51bad0ecaeb59
processor_config.json976 Bece2373c2ae391bce18785a1810543bc6173a1f28b6767cffab2c35dbea5f002
special_tokens_map.json147 KB0a5c981e8c5c6f8886ee007a6d4543a0be6b221cb9ca32a8709384a4c6fc8cbb
tekken.json16.8 MB600bb27946565481ecf51ba8aee252e49b9a68507866080ac9c30185bb312843
tokenizer.json17.1 MBd5f6046775b112f0e2d456ee9dba450684ab964fe5c4e231599bdc6773028135
tokenizer_config.json198 KBf59f7294e4f26383d0ea93840fe21cf197784be0842a8301a0343e8c34ed0d6d
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
b6d637bef2 pinned now manifest updated, same commit2026-11-14 21:24 UTC16 files · 15.4 GB
b6d637bef22026-11-14 21:24 UTC16 files · 15.4 GB

Use it

$ fw get mistralai/Ministral-3-3B-Instruct-2512-BF16
# manifest sha256 f39d12c52f00… · 15.4 GB · 16 files
$ fw verify ./Ministral-3-3B-Instruct-2512-BF16 mistralai/Ministral-3-3B-Instruct-2512-BF16
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/mistralai/Ministral-3-3B-Instruct-2512-BF16.json --dir ./Ministral-3-3B-Instruct-2512-BF16 $ python3 fw.py verify ./Ministral-3-3B-Instruct-2512-BF16 ./mistralai__Ministral-3-3B-Instruct-2512-BF16.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download mistralai/Ministral-3-3B-Instruct-2512-BF16 --revision b6d637bef2393152b3da2b2fde72eecdee30557e --local-dir ./Ministral-3-3B-Instruct-2512-BF16
Run at home

License

Name
Apache License 2.0
Id
Apache-2.0
Class
Permissive
Redistribution
yes
Listing
Hosted (signed)
Decided
License as declared by the publisher. No person read it. 2026-11-14
Text in
README.md
Text sha256
88189c3b964024a8a1d39e0a820596776b997fabb332a151a25973b242f06382
Terms gate
No extra terms beyond the license.

Read the license text

Licenses are checked against the license table. A custom license passes only after a person reads it.

Provenance

From: mistralai <huggingface.co/mistralai>Newsgroups: alt.freeweights.mistralSubject: Ministral-3-3B-Instruct-2512-BF16Date: 2026-11-14Commit: b6d637bSigned: 2DC28CC7B26A283B ✓
Upstream
mistralai/Ministral-3-3B-Instruct-2512-BF16
Pinned commit
b6d637bef2393152b3da2b2fde72eecdee30557e
Manifest
Ministral-3-3B-Instruct-2512-BF16.json · .minisig
Manifest sha256
f39d12c52f0077c81305982c64afe99aedb03aa4a15c83ef3f2e7cac082461bc
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-14 21:24 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/mistralai/Ministral-3-3B-Instruct-2512-BF16.json $ curl -O https://freeweights.allternit.com/manifests/mistralai/Ministral-3-3B-Instruct-2512-BF16.json.minisig $ minisign -Vm Ministral-3-3B-Instruct-2512-BF16.json -P RWQ7KGqyx4zCLXtUFOqvfv8L3g+8fpKJ1FsuxPyOYMF+4KufbUxKq1Oq $ shasum -a 256 <file> # compare with the Files tab

A manifest whose signature fails must be rejected. A file whose size or sha256 does not match must be discarded. Verify a download

Model tree

Parent

mistralai/Ministral-3-3B-Base-2512 · this model is a finetune of it

3 models in the catalog name this one as their parent.

Quantizations
3
Quantizations 3
Get it
Download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download mistralai/Ministral-3-3B-Instruct-2512-BF16
Open in Freeweights Studio

Pull, verify and run it in the Allternit app on Desktop, web or phone.

Signed manifest
minisign, key 2DC28CC7B26A283B. Ministral-3-3B-Instruct-2512-BF16.json. Provenance. Takedown history: none.
Source
huggingface.co/mistralai/Ministral-3-3B-Instruct-2512-BF16

Variants

3 quantized or adapter builds of Ministral-3-3B-Instruct-2512-BF16. Each has its own license check and page.

SubjectSizeParamsLicenseDate
mistralai_Ministral-3-3B-Instruct-2512-GGUFHosted (signed) From: bartowski in alt.freeweights.mistral56.6 GB—Permissive2026-11-14Ministral-3-3B-Instruct-2512-GGUFHosted (signed) From: lmstudio-community in alt.freeweights.mistral9.46 GB—Permissive2026-11-15Ministral-3-3B-Instruct-2512-GGUFHosted (signed) From: ggml-org in alt.freeweights.mistral4.10 GB—Permissive2026-11-15

Related models

All of alt.freeweights.mistral →
From: bartowski <huggingface.co/bartowski>Subject: alamios_Mistral-Small-3.1-DRAFT-0.5B-GGUF
From: microsoft <huggingface.co/microsoft>Subject: chatbench-mistral-7b
From: NousResearch <huggingface.co/NousResearch>Subject: DeepHermes-3-Mistral-24B-Preview
From: mistralai <huggingface.co/mistralai>Subject: Devstral-Small-2-24B-Instruct-2512
From: mlx-community <huggingface.co/mlx-community>Subject: Devstral-Small-2-24B-Instruct-2512-4bit
From: mlx-community <huggingface.co/mlx-community>Subject: Devstral-Small-2-24B-Instruct-2512-6bit