CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.deepseek / DeepSeek-V3.1-4bit
From: mlx-community <huggingface.co/mlx-community>Newsgroups: alt.freeweights.deepseekSubject:

DeepSeek-V3.1-4bit

Date: 2026-11-15Commit: 1634da2Signed: 2DC28CC7B26A283B ✓

Text generation model with 671B parameters. Safetensors weights, 378 GB in all. Published by mlx-community under MIT.

671B params378GBDeepSeek-V3.1-4bitsafetensorspermissive
FactsParams671BSize378 GBFormatssafetensorsTasktext generationLicenseMITFrom sourcepermissive
$ fw get mlx-community/DeepSeek-V3.1-4bit
# manifest sha256 3aef52e0e793… · 378 GB · 99 files
$ fw verify ./DeepSeek-V3.1-4bit mlx-community/DeepSeek-V3.1-4bit
From sourceHosted (signed)mlx-community/DeepSeek-V3.1-4bitLast check: 2026-11-15 05:08 UTC
Get help

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

Quantized of DeepSeek-V3.1.

Works with

Run24Fine-tune29Convert15Evaluate13Scan13Apps58
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
378 GB
Weights
378 GB (safetensors)
Parameters
671B
Formats
safetensors
Task
text generation
License
MIT (permissive)
Files
99, each with a sha256
Publisher
mlx-community
Added
2026-11-15
Section
alt.freeweights.deepseek
Lineage

Original release from the publisher.

Base model: deepseek-ai/DeepSeek-V3.1

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.co2026-11-15 05:08 UTCpassed

Model card

From the publisher at commit 1634da2. Shown as plain text with links; images open at the publisher.

mlx-community/DeepSeek-V3.1-4bit#

This model mlx-community/DeepSeek-V3.1-4bit was converted to MLX format from deepseek-ai/DeepSeek-V3.1 using mlx-lm version 0.26.3.

Use with mlx#

pip install mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/DeepSeek-V3.1-4bit")

prompt = "hello"

if tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)

Files

99 files · 378 GBWeights 378 GB (safetensors). Sizes and sha256 from the signed manifest.
Show all 99 files with their sha256
FileSizesha256
.gitattributes1.52 KB11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361
README.md837 Bb4704566d301369c5636b9451166ed6343137c855a4468fe833277a2c0636445
chat_template.jinja2.92 KB45690185067ea25c073bff2d3e425a45ee2564d2bbf6cdf7d885e139d460b8e0
config.json1.78 KB24e101be55af313a6fa708697e715bf98fdfb633658798d6973c60a3daec1686
configuration_deepseek.py9.90 KBd5126f7a82eaa5fe10c294af7a8f02d59e7145c2dc8837b83ae338c1b72001d2
generation_config.json171 B0ef9febae6b6087f4822b02bc9a1c03a83263dabaa931fb9155d061c8951ca07
model-00001-of-00088.safetensors3.73 GBd2c4b4b1e3dafc828842419be2a15cfcc75e7a266cc5831c22caa4c1fcccd8df
model-00002-of-00088.safetensors4.36 GBed52249e5bbb9d04eda6d2befa00c76022f8c4e28f89110a19525781d703dee3
model-00003-of-00088.safetensors4.23 GBf833308337b38332b1e45883e6119358174039a048747b536cbb9f248d214c2e
model-00004-of-00088.safetensors4.36 GB60385e60f9a90411b3f4dc08d6fc123b3fc551a2981d36549caed39521e895ac
model-00005-of-00088.safetensors4.36 GB1d4527e5195a63bf30cbae38fa7838b1f18899b648cf967fbcca5460c70254da
model-00006-of-00088.safetensors4.23 GB485831b705a609740e8f21bad370b154d9885e5c2f5f38395dd24d098fd9c67c
model-00007-of-00088.safetensors4.36 GB3998ed6febe2cc48dcd8b0a59e39d1bdda41d3da86f4e24def627e42c7104f58
model-00008-of-00088.safetensors4.36 GBc32cd6fd012e04b2198995930ec2f720e552254071b9df8e76bba1271cfba8d7
model-00009-of-00088.safetensors4.23 GBa4b7c632d7e062e5587c0549441d9215eccbec222de116d84659be8b6867e01e
model-00010-of-00088.safetensors4.36 GB081b1e9de8bcc32dd40610a0729847bfa149f2e840c403b3f867cad737a1e1c3
model-00011-of-00088.safetensors4.36 GBc7707cfb6e45b189598b041220bfe31ad764420ca3378e484bcabc6603713998
model-00012-of-00088.safetensors4.23 GB728ad71766d2c04881490dd56d777ba920c38687ffbfbacea706627dafad2b6a
model-00013-of-00088.safetensors4.36 GB88563e2628b43b980038f95996a49f7f4460429b6dcab8d8fd9ae46d8bb28b33
model-00014-of-00088.safetensors4.36 GBfe7267f41fede895a6526f1254100ad1c200abdf7b088ceaaf02ab08166ac8d3
model-00015-of-00088.safetensors4.23 GB91ef3f30f3652938e6ecdacbbc17b25d6c59afaa5ac69c186a132c4e0194e884
model-00016-of-00088.safetensors4.36 GB8876d9cbae74676d7be2b51ccaf8139e246dd52c0fdae6da5cd5928fcb78dd23
model-00017-of-00088.safetensors4.36 GB0135312192c46513b5d659402e93259e7df25efbb0123b1f7b2ba5e816a06091
model-00018-of-00088.safetensors4.23 GB769e9d5192a2f5921ed949b3ef44988de96ea9d6dd055f03a5122efb2b6e3003
model-00019-of-00088.safetensors4.36 GBed734821091dc4a7f92647484169ab4424bccd74ba3db082a7117e302cf37819
model-00020-of-00088.safetensors4.36 GBd305d92e8cd3986bc33bd3b1a76c13476f30a75464ff7725e66a9b4a94fcca97
model-00021-of-00088.safetensors4.23 GB485f7b92bb2bbde1e46a7ab1b3fd73ee53388881bbc85020f83b98ea86789aa7
model-00022-of-00088.safetensors4.36 GB8ccfb8bc9ca485bfd250afc60945176a949577e86c86cd7a7a49da1fb96c58cc
model-00023-of-00088.safetensors4.36 GB3c03082b13ed6ae78d039e246951c7c8ac0c5cd07949a1cc7a6da8b3b9b94930
model-00024-of-00088.safetensors4.23 GBaf059895de124253a152a2612913e85f17dbfc537e60dd32255f52bd675c840f
model-00025-of-00088.safetensors4.36 GB6d0ec20d8cd01ac0dceb68dd81acb39b246334a6e1e54e267b8e77414fb87bb7
model-00026-of-00088.safetensors4.36 GB2cf8d6dda40891d7d341c48dc74c5cadad1c4fc07c522f766f6186db434c5bb6
model-00027-of-00088.safetensors4.23 GB7b64337e05284310704fa27fb0255c61a5158fabd1d4804b30dd6f0186d9c9b7
model-00028-of-00088.safetensors4.36 GBd52d714f1c610ee1b73479927387383c8274afe505ecfaf085906b7e119f64ff
model-00029-of-00088.safetensors4.36 GB3b35cda78a8da1e31985ad2dd9aceca9b00c4fe9578661f0b79d179d18fa3564
model-00030-of-00088.safetensors4.23 GBdb621e7f337f4d34c1e7b5f6fcb90b6c3c7e5a9fb0f61359d7826cf42fd501b2
model-00031-of-00088.safetensors4.36 GBe1c031e0bdd5a31cfc5d7c75100b9f786d112a7b9ed14db1e13f7a920019d196
model-00032-of-00088.safetensors4.36 GB373a599f22bcf5586466a2fb1779097bdefec883606bcff30c9a786b2c91ba46
model-00033-of-00088.safetensors4.23 GB8e7dac3764d00473f511e29f38718d1832fd6839c2341430025ee95c5d1c424d
model-00034-of-00088.safetensors4.36 GB3c156ad7da43da8c30af4fb037d9bdd2007f0292f8edf03fc73a27382b120438
model-00035-of-00088.safetensors4.36 GB68dc640e7f9ae61eb47a173d8b16db9033968fcb3a3141997d2388ec5a326d62
model-00036-of-00088.safetensors4.23 GB29f5b6ed465f7778742310885d70d784167697084d0db274b3de76e78e608aa3
model-00037-of-00088.safetensors4.36 GB1766596e1a6e8d1086ae1b6cc66fd03d6135f3636a9c38c6382e063e548189a1
model-00038-of-00088.safetensors4.36 GBe018356654b0871f9c1386559425841454cf956aaefe7a9a7a5dc26c9fb4004f
model-00039-of-00088.safetensors4.23 GBc2c57d10b95200f227edf5063581645357cbd9a8f68260eec0e48fc5cdef8051
model-00040-of-00088.safetensors4.36 GB580343e99226b6bea8ace1d5539e1b1b57ae79cbdc4cedc9f938b589fbe555ce
model-00041-of-00088.safetensors4.36 GB8d63887b3e4f9b894b1a37ff1bfc62fa9191b7c80e5c78c18eebccbaf6929e7c
model-00042-of-00088.safetensors4.23 GB26a8551bd9b660b619119175705c2c254a6331e2c2f223ef9790318d1b62a0bf
model-00043-of-00088.safetensors4.36 GB840a42a230f4213c9927d548dae6c06652fa7babfdcb26cbfcefc6996b078811
model-00044-of-00088.safetensors4.36 GB8d6dd83f754f9b2993f23446a9f1f2df7ecf200ed6518d8daf9a778f8f7c162a
model-00045-of-00088.safetensors4.23 GB4e195550e412a2b0804e21756b05cc80553687a5524937212f254769c15dbf3a
model-00046-of-00088.safetensors4.36 GB2b285ebfc0adcf008ce3036cf9dc650cfc71836cfabe91d3d652e366927b912f
model-00047-of-00088.safetensors4.36 GB0846b5889de69a9c022b78f679dd1d18474b09fcb9f8093c89f3582c8d511afd
model-00048-of-00088.safetensors4.23 GBbb00e29ec4780f4b1bf7418fa39248fb349b515cc17fbb94cbd908fad3bff56c
model-00049-of-00088.safetensors4.36 GB9c61cc1e5d50d52a6d88fb41a2fac20822f8e2b1ad983fd1c930de206247e4f2
model-00050-of-00088.safetensors4.36 GB6060177879ad2fa3348ba0eb18930e5bbcdf4bada71b035346b11adb1589183b
model-00051-of-00088.safetensors4.23 GBc37bcd50507df4d1cbc92a42819f7deae3850ddca0fc868e8bb3c91cd04581c3
model-00052-of-00088.safetensors4.36 GB3c699960fe5f86f8a0bd851048488f6e3d196f666ffe1275f2efcb7900b15325
model-00053-of-00088.safetensors4.36 GB5eb41d34a8a91646a0fb2b006efdcb5fb0e03cbada10cb1a538660edca936235
model-00054-of-00088.safetensors4.23 GB4b3e8acf4a5761be48e2fce23382c32c55f7fd14b419032a531bb08c0322a0eb
model-00055-of-00088.safetensors4.36 GB4d75a5d6cfeababd09763e8cdf48c00a58633f42dc784f8a76f144bb91f19978
model-00056-of-00088.safetensors4.36 GBc164fec9f9c6124541a8453d25df0e72911038338535664e0d02b686b535cd05
model-00057-of-00088.safetensors4.23 GB0db61034e5b5ec3494823e7ab6fd117bc57c82f67bc2d0f8c5d23da80573a10d
model-00058-of-00088.safetensors4.36 GBa14c079a7ffb323559a01084fade866e7a5afbce0701e156f6fd6d221aecb926
model-00059-of-00088.safetensors4.36 GBbacdc8abd8eb90107321e2d03d46215209c0854c168d5dd17c5d4a5e7a1ad06c
model-00060-of-00088.safetensors4.23 GB8dcf83663b05807d269744e91a66dabb4a93d6892869b4baed820c7a8c82995f
model-00061-of-00088.safetensors4.36 GB44fae762b44dfdf7001a785c8e8af23f97d0070fec5df01d6d2517ca9945f944
model-00062-of-00088.safetensors4.36 GBb04e95f752fd3693a9fdeafa3adad4006634d2edd3fb805c925c923fb76eb9dc
model-00063-of-00088.safetensors4.23 GB7a31117b0b3d5b7f63e9cf35292a739a52ebba04f1bcf582b0b9cd4f680e87f9
model-00064-of-00088.safetensors4.36 GBf0f57186eb7e059c6bbfaf25c1936e8b91b4f181fa43f1a6bb40802540458c91
model-00065-of-00088.safetensors4.36 GB85fc61fc0dc8e70092a0bdac1bea01587ea6f7c1809a26d4b1c15630a38c20ff
model-00066-of-00088.safetensors4.23 GB79b297ca2c4f80cc688422a03bf02c72bbbf7248d51a9116a12facd603e67bbf
model-00067-of-00088.safetensors4.36 GB72f7d6bc8f5c86b2d45a45760b923d2b9bb5af7ecee0990d1440e143e257301e
model-00068-of-00088.safetensors4.36 GB1177f33114ec600525fd6283f920ac68bbcfdc6d16165c0278b3d25950ea9955
model-00069-of-00088.safetensors4.23 GB0b60659e17d83cf89a681d0a93e5ebf96cfd06417766bf8f3860800e79d0c0f7
model-00070-of-00088.safetensors4.36 GB751be31e1a7badb4f18483fcace26e682f224334f224545aa8baf7cbc0acbd8f
model-00071-of-00088.safetensors4.36 GB3b908ef007918f7026e686a278beb2ea7ad8b1dc15f6cf9a12542830a9d81f51
model-00072-of-00088.safetensors4.23 GB02f98a537f88cf7a5620ba367a62e55b8e0e90c316d80fb4010037edba05cd97
model-00073-of-00088.safetensors4.36 GB8444963d4a9e0ef68fb06740fd84f713afc2e809e50e3577d3a9de147f6abbcd
model-00074-of-00088.safetensors4.36 GB04f1ab35d9a8563773ebfb88b02f24c675a8802c46436991db23c6285e540771
model-00075-of-00088.safetensors4.23 GB861bf317e5563e413fe9dec23f0c9e0e6299372f50643283162bb502b80f311d
model-00076-of-00088.safetensors4.36 GB75c6597782b745043036353c82c0a33f7aa24af537634e2de6427c602765b9a2
model-00077-of-00088.safetensors4.36 GB5a516461d6796bc316005008c8e6eef6327c34eab65d8077a03c6583a832b61c
model-00078-of-00088.safetensors4.23 GB44806857475acc57dcdb1663fcbbd0b8794f299e60c3c6b22a143d444e66a8ac
model-00079-of-00088.safetensors4.36 GBd4849aac66d21407b01eefd6633440885d85276091766370146d65f1e7da066f
model-00080-of-00088.safetensors4.36 GB3c6eb877889afc09b06473672e84c9d64bd69583fae293ddbd58f1592b51a36d
model-00081-of-00088.safetensors4.23 GB138ad9b2ab3d459c8fbd42793d48a571012c40e057f207d8137ae7ee587405cd
model-00082-of-00088.safetensors4.36 GBd69af5f1ba5c883f4c20412e04f28cfaf28643b4f3d415582ea50bead8b6be5d
model-00083-of-00088.safetensors4.36 GB5b78071da5c8a69c25a06f392377de16bc8a62d6c9cced7601713ee20c12c104
model-00084-of-00088.safetensors4.23 GBbe421f0a9cc9c0b069cd6a20ee8927868285550fefe3b58183396bb37c9505d8
model-00085-of-00088.safetensors4.36 GB37e33be0a0f0a7c4c9afcd70e68a9bd70db59dd3069ed97eb578a4db959dcfa2
model-00086-of-00088.safetensors4.36 GB75efd61da99d898acb8146d6925dd546a3a8ee9c329af75aebe59b663e2b1703
model-00087-of-00088.safetensors4.23 GBc323d052bea6e0aad3dc5b57cf6dbf05fd5dfa4852d57cf685f5e0d9720296c0
model-00088-of-00088.safetensors2.66 GB1136d19be2ccbd32096a3a5e957df175cf769f19a83c1bade033816c00639da2
model.safetensors.index.json219 KB4f4ba383f06d90a96b567919e62eb26dba07964b2da2555bba34550e1f9178b4
modeling_deepseek.py75.7 KBeb6d535742061cf9da7a4232ecfe2f089d5ce20080afc5d2181e9d70ec4af8a4
special_tokens_map.json485 B59cda48bbe8bab9d61ffb410e6e3c07b6d98bff73cee7c88ff8b51f95f21ab1c
tokenizer.json9.98 MBa59031bae85a334b70c9f91582c76b3b11a9c7f7500c62a820f3f9433f92e877
tokenizer_config.json164 KB20ad3e88e91770f5cb56f0abdec2082d96081c0e5e1b676742cb94a31e9ef60a
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
1634da2179 pinned now2026-11-15 05:08 UTC99 files · 378 GB

Use it

$ fw get mlx-community/DeepSeek-V3.1-4bit
# manifest sha256 3aef52e0e793… · 378 GB · 99 files
$ fw verify ./DeepSeek-V3.1-4bit mlx-community/DeepSeek-V3.1-4bit
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/mlx-community/DeepSeek-V3.1-4bit.json --dir ./DeepSeek-V3.1-4bit $ python3 fw.py verify ./DeepSeek-V3.1-4bit ./mlx-community__DeepSeek-V3.1-4bit.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download mlx-community/DeepSeek-V3.1-4bit --revision 1634da2179770a14024405afa6d1e0ce70a71ff4 --local-dir ./DeepSeek-V3.1-4bit
transformersLoad the weights in Python.
import os os.environ["HF_ENDPOINT"] = "https://freeweights.allternit.com" from transformers import AutoModelForCausalLM, AutoTokenizer tok = AutoTokenizer.from_pretrained("mlx-community/DeepSeek-V3.1-4bit", revision="1634da2179770a14024405afa6d1e0ce70a71ff4") model = AutoModelForCausalLM.from_pretrained("mlx-community/DeepSeek-V3.1-4bit", revision="1634da2179770a14024405afa6d1e0ce70a71ff4", device_map="auto")
vLLMServe an OpenAI-compatible API.
$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve mlx-community/DeepSeek-V3.1-4bit --revision 1634da2179770a14024405afa6d1e0ce70a71ff4

License

Name
MIT License
Id
MIT
Class
Permissive
Redistribution
yes
Listing
Hosted (signed)
Decided
License as declared by the publisher. No person read it. 2026-11-15
Text in
README.md
Text sha256
b4704566d301369c5636b9451166ed6343137c855a4468fe833277a2c0636445
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: mlx-community <huggingface.co/mlx-community>Newsgroups: alt.freeweights.deepseekSubject: DeepSeek-V3.1-4bitDate: 2026-11-15Commit: 1634da2Signed: 2DC28CC7B26A283B ✓
Upstream
mlx-community/DeepSeek-V3.1-4bit
Pinned commit
1634da2179770a14024405afa6d1e0ce70a71ff4
Manifest
DeepSeek-V3.1-4bit.json · .minisig
Manifest sha256
3aef52e0e793eacf51323d1474a2ce78777a4ded40931c5ea730c26b2e8242e6
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 05:08 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/mlx-community/DeepSeek-V3.1-4bit.json $ curl -O https://freeweights.allternit.com/manifests/mlx-community/DeepSeek-V3.1-4bit.json.minisig $ minisign -Vm DeepSeek-V3.1-4bit.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

deepseek-ai/DeepSeek-V3.1 · this model is a quantization of it

No model in the catalog names this one as its parent.

Get it
Download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download mlx-community/DeepSeek-V3.1-4bit
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. DeepSeek-V3.1-4bit.json. Provenance. Takedown history: none.
Source
huggingface.co/mlx-community/DeepSeek-V3.1-4bit

Related models

All of alt.freeweights.deepseek →
From: timm <huggingface.co/timm>Subject: deepseek_vit_412m.deepseek_v4_1_flash
From: bartowski <huggingface.co/bartowski>Subject: deepseek-ai_DeepSeek-R1-0528-GGUF
From: bartowski <huggingface.co/bartowski>Subject: deepseek-ai_DeepSeek-R1-0528-Qwen3-8B-GGUF
From: bartowski <huggingface.co/bartowski>Subject: deepseek-ai_DeepSeek-V3-0324-GGUF
From: bartowski <huggingface.co/bartowski>Subject: deepseek-ai_DeepSeek-V3.1-GGUF
From: mlx-community <huggingface.co/mlx-community>Subject: deepseek-ai-DeepSeek-V4-Flash-8bit