CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.deepseek / deepseek-ai-DeepSeek-V4-Flash-8bit
From: mlx-community <huggingface.co/mlx-community>Newsgroups: alt.freeweights.deepseekSubject:

deepseek-ai-DeepSeek-V4-Flash-8bit

Date: 2026-11-15Commit: 72807dcSigned: 2DC28CC7B26A283B ✓

Text generation model with 284.3B parameters. Safetensors weights, 302 GB in all. Published by mlx-community under MIT.

284.3B params302GBdeepseek-ai-DeepSeek-V4-Flash-8bitsafetensorspermissive
FactsParams284.3BSize302 GBFormatssafetensorsTasktext generationLicenseMITFrom sourcepermissive
$ fw get mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit
# manifest sha256 f29d127e0f0d… · 302 GB · 72 files
$ fw verify ./deepseek-ai-DeepSeek-V4-Flash-8bit mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit
From sourceHosted (signed)mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bitLast check: 2026-11-15 05:14 UTC
Get help

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

Quantized of DeepSeek-V4-Flash.

Works with

Run24Fine-tune29Convert15Evaluate13Scan13Apps58
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
302 GB
Weights
302 GB (safetensors)
Parameters
284.3B
Formats
safetensors
Task
text generation
License
MIT (permissive)
Files
72, 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-V4-Flash

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 05:14 UTCpassed

Model card

From the publisher at commit 72807dc. Shown as plain text with links; images open at the publisher.

mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit#

This model mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit was converted to MLX format from deepseek-ai/DeepSeek-V4-Flash using mlx-lm version 0.31.3.

Use with mlx#

pip install mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit")

prompt = "hello"

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

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

Files

72 files · 302 GBWeights 302 GB (safetensors). Sizes and sha256 from the signed manifest.
Show all 72 files with their sha256
FileSizesha256
.gitattributes1.52 KB11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361
README.md933 B15539873844343e655fb0d9fd27d99d556947b7e025f1c4c75351e464130d37a
config.json2.21 KB7f49d78c9654babfa0d8218eb0f8d3c1a6ff893782bcad774a1481fa2c98d336
generation_config.json170 B5fccff80f55a4d455bbe516bdd552edf3e9623df95e99fbf2a3c3389fdf91af0
model-00001-of-00065.safetensors5.25 GB7c0980e7b53ac3c6e9c4063ae55794729d568fb8e98f382307712d2fa17b6c2b
model-00002-of-00065.safetensors4.72 GBf92fa6509012525ac8a593147bef95055d68aa3a1c1beebbcea79226fdccf822
model-00003-of-00065.safetensors4.74 GBa82c53c4efadff140981d318ff3611df6f29d72bb866e98796720c159efa7165
model-00004-of-00065.safetensors4.56 GB9f235c15e314cfa1445cf26031cf7d5d2840d0f9826424678f0bf532d9d25184
model-00005-of-00065.safetensors4.71 GB7a1726e111d7fcd8df5595d6ed53632d15070646ce9dba198fe1575b5cf6a5b7
model-00006-of-00065.safetensors4.73 GBa342a57752ba2ffd965d49518d88e91f6bacf7b785c13428a67c99e2236128b5
model-00007-of-00065.safetensors4.56 GBf680a04cdcb18fed48d39d1c32262a230a13757b33d92a5c129d6aed20b67005
model-00008-of-00065.safetensors4.71 GB6b432d71cea7cb5732a756dd84151fd24ba1c7406665539b3f82d86fa7b1cde6
model-00009-of-00065.safetensors4.73 GB48c0c47212ea6ed291bf7da9018c56f0ce37551c079b1dc7ac7a7b1f0934d3c7
model-00010-of-00065.safetensors4.56 GB2dda5e5dcc64a1aeafa7d924941e187a66355490ab6179a0676b4f3baa9dc9d6
model-00011-of-00065.safetensors4.71 GB4eace491a28b37c625f6f272551f2ac75a23fa5ffb6e4be087966a15b19385f0
model-00012-of-00065.safetensors4.73 GBbf950b3eec3dd6b87e6a5b385b9c897c4fdc222e4189ffe968246059c3731c81
model-00013-of-00065.safetensors4.56 GB602baa32208adb814fe845565889b57456001af35430f38a37292f21710f13bd
model-00014-of-00065.safetensors4.71 GB714aaa557bc1c6936a6cd11f9d4e2ccfb5905ee9b01b5f3a07c7601eaf325b2c
model-00015-of-00065.safetensors4.73 GB7b15eccd92548c10e3526301ef4a69918c96c81bd543be791b8c773ab65ef37d
model-00016-of-00065.safetensors4.56 GB2dd687eb75a8f597fe31a07587c36177c543767da84428cf31b016a8492644d0
model-00017-of-00065.safetensors4.71 GBa645b53b07b37774d0782e51a162c5bb3481079adb943bd73038cd444a3153a4
model-00018-of-00065.safetensors4.73 GB024a025219c7bc13e59fb97c8e612c9674a73147a90c2c4d62b4482554ec90a1
model-00019-of-00065.safetensors4.56 GBf475b8d84471313edfdaefeb0771ea4a72f04ca4c2f6ef250ad6d0d27ecde3d3
model-00020-of-00065.safetensors4.71 GBde98de122328c0025564b3b3792251a55738be8d568e6695843707c29d09dd7e
model-00021-of-00065.safetensors4.73 GB7e5030ce5cfdd9538ccad72ae7d93b8a237b6f010150eeda99cf9019105dee27
model-00022-of-00065.safetensors4.56 GB499f0e6c0a17af6323a608bd327f105c7d4c4e66cbac6623cbe5675fd620f24d
model-00023-of-00065.safetensors4.71 GBf56c834e51bb29d3a2930c1bf1339ff24204a3a1678a1ed145dcfaa08dc9a512
model-00024-of-00065.safetensors4.73 GB8baa06362b3df2acecb8ca3cfc1d854c1b33bf3da9b94d1d68d077a238864e61
model-00025-of-00065.safetensors4.56 GB2428eb4a09b75efd86f89fa59bed4f13d13eb99faff4f8714b38e9d52fd8c9c9
model-00026-of-00065.safetensors4.71 GBfb754a30ed907316231edd5b905afcd226b44ec78637ab1ad50b9e5f7a845683
model-00027-of-00065.safetensors4.73 GB793a1e59363ddb5ba8a6df29e00a7cc656cc1a1dc2d28584831c96c2d36e397d
model-00028-of-00065.safetensors4.56 GBf66064a778323600485a985219b37ef6af5030b6b5086e86faf00ae5019916df
model-00029-of-00065.safetensors4.71 GBfccb5aa3da1963642ff5d2c4f1fa31b82109293bdd477345ead8349bbe1d9cb8
model-00030-of-00065.safetensors4.73 GBd55f381dc85b6034dc068de375739b1edca64c51e21d4aac690b385de552d67c
model-00031-of-00065.safetensors4.56 GBe19b58624726290af07de1a31db0b7c35dee22f4ab05f686b457e7feab77d0de
model-00032-of-00065.safetensors4.71 GB9f78c277c4586a5c71d7ef715c5b07643e9a3404d7b2d81ef23d2e805752d9d3
model-00033-of-00065.safetensors4.73 GBab12825a07b17eded0858efeb1b97fd32a301f0e549b4997e2fb16bca4ffbde4
model-00034-of-00065.safetensors4.56 GB9eb7275b9cdea1e02f92ebbddecb1b7e2c6f22e2da34647b30d5bb55c14be6e0
model-00035-of-00065.safetensors4.71 GB9685230ff4cd5b89340f100a7e99cd4563b34c65ee4f03a29eb8b1af44937bf1
model-00036-of-00065.safetensors4.73 GBa36d3964cd97acd0ca133bc2237f01a9b13f87df042f5a0745bc8a1fcaf2cd09
model-00037-of-00065.safetensors4.56 GB50e2c760fd2b10a785a990b7539d811ae4a10b01ae77c67789a7fefabb2e0078
model-00038-of-00065.safetensors4.71 GBda3cc38df631dd1fa74bc9d24671014e095eebe3045387ecc418fba551197a31
model-00039-of-00065.safetensors4.73 GB481690ad72daff6ad1dddbfa42b013bfcfce2d6c661d6e7a970c4cbc35bb5c32
model-00040-of-00065.safetensors4.56 GBd90633f9227aeb2038d4a6f358d092cef1b4a9df7f52436201cee24fffa605c8
model-00041-of-00065.safetensors4.71 GB69db8ba1c192faa31333949197fad4895900397c1e9cf40cdb1432891ded2932
model-00042-of-00065.safetensors4.73 GBc0c2447301e297e8667f2420b93e1905f658a3706918165ee4c2c20d6f0a713d
model-00043-of-00065.safetensors4.56 GBc51c25159c0c33b78054921333c05b5486df07c583b6ba18af08c0d6e24fcadd
model-00044-of-00065.safetensors4.71 GB02369afd436852eff3258f6cad9bd2c8770d567128d832df2e68a59463fb600a
model-00045-of-00065.safetensors4.73 GB0699f846e1aa16fce67a2b68538218c6ec74e5ecad24e8705913bf8dc3ea48b0
model-00046-of-00065.safetensors4.56 GB0503aa1eb1f978058474a14aa7a34cb4f357f681fe317ea6dc89da1460393ae4
model-00047-of-00065.safetensors4.71 GB3f903c7ab2b1479c95a4977630ec08c92864bfec7d6e54fff8a6ef34fb5a83cf
model-00048-of-00065.safetensors4.73 GBe3ff4ed7c47690a86c0be083f83b8c238979298c75fff64e082b542f6965f074
model-00049-of-00065.safetensors4.56 GB31f5df2ad7c3ed5a02dae04bc74d6c54998e7559c9112ef0c46b1dbe011020dc
model-00050-of-00065.safetensors4.71 GB10fda7f48e0467321524161d509398bcc292c18ad118fda250fd935ef89bf903
model-00051-of-00065.safetensors4.73 GBba773a34be85f22b430d07f5120f607959091cbbe027290a4ef97b571c036692
model-00052-of-00065.safetensors4.56 GBca5ef30db44db7752b601000932526d505799a3e55c01781caffc0426bedce91
model-00053-of-00065.safetensors4.71 GB68a2952c2e9c80babeb59dc3bc30518a999b6ef610e7257b315987c069341c92
model-00054-of-00065.safetensors4.73 GBdac2d764b55d06f0b38bcc45b13b035b519327835268f5168b4168b38c378aff
model-00055-of-00065.safetensors4.56 GBbfaf479ad6af7c1dd2e0bb49a1e06d28f461d4715f6af590a2741fe656cd284f
model-00056-of-00065.safetensors4.71 GB32daf334c22a8a7e7c81340ed05ceaeb7aed5960ecf3b4062ff0cde05f0258f6
model-00057-of-00065.safetensors4.73 GB3cffc0b84cc864be1b633cff03415ed363232dc21db55a27150d4b46beb52a9a
model-00058-of-00065.safetensors4.56 GB0c8bd1650f75e0e36232db11061e71cf053e36638d69c6c606d7d49fcfba147b
model-00059-of-00065.safetensors4.71 GB74c14a310e31812f2609b9f4039557f7f35ee1d9ee6fcd3d41a85043fb5a66d1
model-00060-of-00065.safetensors4.73 GBc1189ed9dc15a2d28f3910bc7e8aec2c969a5692d5b35d1bc5ca8dd59c83447e
model-00061-of-00065.safetensors4.56 GBe0f0002d3d4a0c6e3cec6cb6d99b7dd7cda99bb6fdcd6f93718e6b8641d25ad7
model-00062-of-00065.safetensors4.71 GBf237c985e07165a15f0e61e262000c3c482efc1aa2de5cb629bf18e499c2493e
model-00063-of-00065.safetensors4.73 GBfa88bc8a1d84c20416d6e045c6e0d6294919315eb05f55c58a37907bd60fb503
model-00064-of-00065.safetensors4.56 GB4e536963d6edb054faf2278e56d452c83ed971ccfd297004d243ed010130667a
model-00065-of-00065.safetensors2.87 GB40b2422035b211675e97f31fa4913ac5fcc20ec084808135b577fd975c81cc8c
model.safetensors.index.json274 KB4c2a33e508c7cc0d714e64b7a1b3ebf0f23c11c0c801ecaf4283b250d67bc18b
tokenizer.json10.1 MB8421075aceec0741baf15179b6c337f443f9029e53ea88d8c1e909dd0d69b70b
tokenizer_config.json397 B21aa165d554e2c2f252a49730b71962edc15c36bbfa4b090ab7bd3340a2fa549
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
72807dce70 pinned now2026-11-15 05:14 UTC72 files · 302 GB

Use it

$ fw get mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit
# manifest sha256 f29d127e0f0d… · 302 GB · 72 files
$ fw verify ./deepseek-ai-DeepSeek-V4-Flash-8bit mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit
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-ai-DeepSeek-V4-Flash-8bit.json --dir ./deepseek-ai-DeepSeek-V4-Flash-8bit $ python3 fw.py verify ./deepseek-ai-DeepSeek-V4-Flash-8bit ./mlx-community__deepseek-ai-DeepSeek-V4-Flash-8bit.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit --revision 72807dce705791c8dcc4cb8f966dcb6f2207b558 --local-dir ./deepseek-ai-DeepSeek-V4-Flash-8bit
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-ai-DeepSeek-V4-Flash-8bit", revision="72807dce705791c8dcc4cb8f966dcb6f2207b558") model = AutoModelForCausalLM.from_pretrained("mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit", revision="72807dce705791c8dcc4cb8f966dcb6f2207b558", device_map="auto")
vLLMServe an OpenAI-compatible API.
$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit --revision 72807dce705791c8dcc4cb8f966dcb6f2207b558

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
15539873844343e655fb0d9fd27d99d556947b7e025f1c4c75351e464130d37a
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-ai-DeepSeek-V4-Flash-8bitDate: 2026-11-15Commit: 72807dcSigned: 2DC28CC7B26A283B ✓
Upstream
mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit
Pinned commit
72807dce705791c8dcc4cb8f966dcb6f2207b558
Manifest
deepseek-ai-DeepSeek-V4-Flash-8bit.json · .minisig
Manifest sha256
f29d127e0f0d7f2c0154df57da4688cee76c4131ea0cc54c91749810d344e8cd
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 05:14 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit.json $ curl -O https://freeweights.allternit.com/manifests/mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit.json.minisig $ minisign -Vm deepseek-ai-DeepSeek-V4-Flash-8bit.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-V4-Flash · 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-ai-DeepSeek-V4-Flash-8bit
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. deepseek-ai-DeepSeek-V4-Flash-8bit.json. Provenance. Takedown history: none.
Source
huggingface.co/mlx-community/deepseek-ai-DeepSeek-V4-Flash-8bit

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-OCR-8bit