CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.deepseek / DeepSeek-R1-Distill-Qwen-32B-GGUF
From: bartowski <huggingface.co/bartowski>Newsgroups: alt.freeweights.deepseekSubject:

DeepSeek-R1-Distill-Qwen-32B-GGUF

Date: 2026-11-14Commit: 1dc8cf9Signed: 2DC28CC7B26A283B ✓

Text generation model. Gguf weights, 519 GB in all. Published by bartowski under MIT.

deepseek519GBDeepSeek-R1-Distill-Qwen-32B-GGUFggufpermissive
FactsParamsnot statedSize519 GBFormatsggufTasktext generationLicenseMITFrom sourcepermissive
$ fw get bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF
# manifest sha256 29990db1c5b9… · 519 GB · 30 files
$ fw verify ./DeepSeek-R1-Distill-Qwen-32B-GGUF bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF
From sourceHosted (signed)bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUFLast check: 2026-11-14 21:27 UTC
Run at homeGet help

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

Quantized of DeepSeek-R1-Distill-Qwen-32B.

Works with

Run14Convert1Evaluate13Scan12Apps58

Convert with

ggml
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
519 GB
Weights
9.03 GB (smallest of 26 GGUF variants)
Parameters
not stated
Formats
gguf
Task
text generation
License
MIT (permissive)
Files
30, each with a sha256
Publisher
bartowski
Added
2026-11-14
Section
alt.freeweights.deepseek
Lineage

Original release from the publisher.

Base model: deepseek-ai/DeepSeek-R1-Distill-Qwen-32B

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 21:27 UTCpassed

Model card

From the publisher at commit 1dc8cf9. Shown as plain text with links; images open at the publisher.

Llamacpp imatrix Quantizations of DeepSeek-R1-Distill-Qwen-32B#

Using llama.cpp release b4514 for quantization.

Original model: https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B

All quants made using imatrix option with dataset from here

Run them in LM Studio

Prompt format#

<｜begin▁of▁sentence｜>{system_prompt}<｜User｜>{prompt}<｜Assistant｜>

Download a file (not the whole branch) from below:#

FilenameQuant typeFile SizeSplitDescription
DeepSeek-R1-Distill-Qwen-32B-bf16.ggufbf1665.54GBtrueFull BF16 weights.
DeepSeek-R1-Distill-Qwen-32B-Q8_0.ggufQ8_034.82GBfalseExtremely high quality, generally unneeded but max available quant.
DeepSeek-R1-Distill-Qwen-32B-Q6_K_L.ggufQ6_K_L27.26GBfalseUses Q8_0 for embed and output weights. Very high quality, near perfect, recommended.
DeepSeek-R1-Distill-Qwen-32B-Q6_K.ggufQ6_K26.89GBfalseVery high quality, near perfect, recommended.
DeepSeek-R1-Distill-Qwen-32B-Q5_K_L.ggufQ5_K_L23.74GBfalseUses Q8_0 for embed and output weights. High quality, recommended.
DeepSeek-R1-Distill-Qwen-32B-Q5_K_M.ggufQ5_K_M23.26GBfalseHigh quality, recommended.
DeepSeek-R1-Distill-Qwen-32B-Q5_K_S.ggufQ5_K_S22.64GBfalseHigh quality, recommended.
DeepSeek-R1-Distill-Qwen-32B-Q4_1.ggufQ4_120.64GBfalseLegacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon.
DeepSeek-R1-Distill-Qwen-32B-Q4_K_L.ggufQ4_K_L20.43GBfalseUses Q8_0 for embed and output weights. Good quality, recommended.
DeepSeek-R1-Distill-Qwen-32B-Q4_K_M.ggufQ4_K_M19.85GBfalseGood quality, default size for most use cases, recommended.
DeepSeek-R1-Distill-Qwen-32B-Q4_K_S.ggufQ4_K_S18.78GBfalseSlightly lower quality with more space savings, recommended.
DeepSeek-R1-Distill-Qwen-32B-Q4_0.ggufQ4_018.71GBfalseLegacy format, offers online repacking for ARM and AVX CPU inference.
DeepSeek-R1-Distill-Qwen-32B-IQ4_NL.ggufIQ4_NL18.68GBfalseSimilar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference.
DeepSeek-R1-Distill-Qwen-32B-Q3_K_XL.ggufQ3_K_XL17.93GBfalseUses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability.
DeepSeek-R1-Distill-Qwen-32B-IQ4_XS.ggufIQ4_XS17.69GBfalseDecent quality, smaller than Q4_K_S with similar performance, recommended.
DeepSeek-R1-Distill-Qwen-32B-Q3_K_L.ggufQ3_K_L17.25GBfalseLower quality but usable, good for low RAM availability.
DeepSeek-R1-Distill-Qwen-32B-Q3_K_M.ggufQ3_K_M15.94GBfalseLow quality.
DeepSeek-R1-Distill-Qwen-32B-IQ3_M.ggufIQ3_M14.81GBfalseMedium-low quality, new method with decent performance comparable to Q3_K_M.
DeepSeek-R1-Distill-Qwen-32B-Q3_K_S.ggufQ3_K_S14.39GBfalseLow quality, not recommended.
DeepSeek-R1-Distill-Qwen-32B-IQ3_XS.ggufIQ3_XS13.71GBfalseLower quality, new method with decent performance, slightly better than Q3_K_S.
DeepSeek-R1-Distill-Qwen-32B-Q2_K_L.ggufQ2_K_L13.07GBfalseUses Q8_0 for embed and output weights. Very low quality but surprisingly usable.
DeepSeek-R1-Distill-Qwen-32B-Q2_K.ggufQ2_K12.31GBfalseVery low quality but surprisingly usable.
DeepSeek-R1-Distill-Qwen-32B-IQ2_M.ggufIQ2_M11.26GBfalseRelatively low quality, uses SOTA techniques to be surprisingly usable.
DeepSeek-R1-Distill-Qwen-32B-IQ2_S.ggufIQ2_S10.39GBfalseLow quality, uses SOTA techniques to be usable.
DeepSeek-R1-Distill-Qwen-32B-IQ2_XS.ggufIQ2_XS9.96GBfalseLow quality, uses SOTA techniques to be usable.
DeepSeek-R1-Distill-Qwen-32B-IQ2_XXS.ggufIQ2_XXS9.03GBfalseVery low quality, uses SOTA techniques to be usable.

Embed/output weights#

Some of these quants (Q3_K_XL, Q4_K_L etc) are the standard quantization method with the embeddings and output weights quantized to Q8_0 instead of what they would normally default to.

Downloading using huggingface-cli#

Click to view download instructions

First, make sure you have hugginface-cli installed:

pip install -U "huggingface_hub[cli]"

Then, you can target the specific file you want:

huggingface-cli download bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF --include "DeepSeek-R1-Distill-Qwen-32B-Q4_K_M.gguf" --local-dir ./

If the model is bigger than 50GB, it will have been split into multiple files. In order to download them all to a local folder, run:

huggingface-cli download bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF --include "DeepSeek-R1-Distill-Qwen-32B-Q8_0/*" --local-dir ./

You can either specify a new local-dir (DeepSeek-R1-Distill-Qwen-32B-Q8_0) or download them all in place (./)

ARM/AVX information#

Previously, you would download Q4_0_4_4/4_8/8_8, and these would have their weights interleaved in memory in order to improve performance on ARM and AVX machines by loading up more data in one pass.

Now, however, there is something called "online repacking" for weights. details in this PR. If you use Q4_0 and your hardware would benefit from repacking weights, it will do it automatically on the fly.

As of llama.cpp build b4282 you will not be able to run the Q4_0_X_X files and will instead need to use Q4_0.

Additionally, if you want to get slightly better quality for , you can use IQ4_NL thanks to this PR which will also repack the weights for ARM, though only the 4_4 for now. The loading time may be slower but it will result in an overall speed incrase.

Click to view Q4_0_X_X information (deprecated

I'm keeping this section to show the potential theoretical uplift in performance from using the Q4_0 with online repacking.

Click to view benchmarks on an AVX2 system (EPYC7702)

modelsizeparamsbackendthreadstestt/s% (vs Q4_0)
qwen2 3B Q4_01.70 GiB3.09 BCPU64pp512204.03 ± 1.03100%
qwen2 3B Q4_01.70 GiB3.09 BCPU64pp1024282.92 ± 0.19100%
qwen2 3B Q4_01.70 GiB3.09 BCPU64pp2048259.49 ± 0.44100%
qwen2 3B Q4_01.70 GiB3.09 BCPU64tg12839.12 ± 0.27100%
qwen2 3B Q4_01.70 GiB3.09 BCPU64tg25639.31 ± 0.69100%
qwen2 3B Q4_01.70 GiB3.09 BCPU64tg51240.52 ± 0.03100%
qwen2 3B Q4_K_M1.79 GiB3.09 BCPU64pp512301.02 ± 1.74147%
qwen2 3B Q4_K_M1.79 GiB3.09 BCPU64pp1024287.23 ± 0.20101%
qwen2 3B Q4_K_M1.79 GiB3.09 BCPU64pp2048262.77 ± 1.81101%
qwen2 3B Q4_K_M1.79 GiB3.09 BCPU64tg12818.80 ± 0.9948%
qwen2 3B Q4_K_M1.79 GiB3.09 BCPU64tg25624.46 ± 3.0483%
qwen2 3B Q4_K_M1.79 GiB3.09 BCPU64tg51236.32 ± 3.5990%
qwen2 3B Q4_0_8_81.69 GiB3.09 BCPU64pp512271.71 ± 3.53133%
qwen2 3B Q4_0_8_81.69 GiB3.09 BCPU64pp1024279.86 ± 45.63100%
qwen2 3B Q4_0_8_81.69 GiB3.09 BCPU64pp2048320.77 ± 5.00124%
qwen2 3B Q4_0_8_81.69 GiB3.09 BCPU64tg12843.51 ± 0.05111%
qwen2 3B Q4_0_8_81.69 GiB3.09 BCPU64tg25643.35 ± 0.09110%
qwen2 3B Q4_0_8_81.69 GiB3.09 BCPU64tg51242.60 ± 0.31105%

Q4_0_8_8 offers a nice bump to prompt processing and a small bump to text generation

Which file should I choose?#

Click here for details

A great write up with charts showing various performances is provided by Artefact2 here

The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.

If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.

If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.

Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.

If you don't want to think too much, grab one of the K-quants. These are in format 'QX_K_X', like Q5_K_M.

If you want to get more into the weeds, you can check out this extremely useful feature chart:

llama.cpp feature matrix

But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQX_X, like IQ3_M. These are newer and offer better performance for their size.

These I-quants can also be used on CPU and Apple Metal, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.

The I-quants are not compatible with Vulcan, which is also AMD, so if you have an AMD card double check if you're using the rocBLAS build or the Vulcan build. At the time of writing this, LM Studio has a preview with ROCm support, and other inference engines have specific builds for ROCm.

Credits#

Thank you kalomaze and Dampf for assistance in creating the imatrix calibration dataset.

Thank you ZeroWw for the inspiration to experiment with embed/output.

Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski

Files

30 files · 519 GBWeights 9.03 GB (smallest of 26 GGUF variants). Sizes and sha256 from the signed manifest.
Show all 30 files with their sha256
FileSizesha256
.gitattributes3.75 KB2744ba033b366a24e08b6f34a1e2d21f0299643373142356546eec9dc6268abf
DeepSeek-R1-Distill-Qwen-32B-IQ2_M.gguf11.3 GBfe09c039ff7a9dbb4b45c2d69117015320eca143438c9e3edb1717631c4facfc
DeepSeek-R1-Distill-Qwen-32B-IQ2_S.gguf10.4 GBb35c64a328cb754fda5d758147642e19b7c4c86f82a245dd0f2df95c1a791623
DeepSeek-R1-Distill-Qwen-32B-IQ2_XS.gguf9.96 GBa8168ec28aa6d9b4d3d0b4980dba3d081c7677840163c1f85e9336ccc3d8a4f0
DeepSeek-R1-Distill-Qwen-32B-IQ2_XXS.gguf9.03 GB8b0c65db4a9ecc9e07eb1cd524a32d4e9f8c2cd95cdc37f76c62e0999ea3e999
DeepSeek-R1-Distill-Qwen-32B-IQ3_M.gguf14.8 GBda94a7c4b8f37af1cfeba68279df713b01d200476a77db9244a2e5ade1347175
DeepSeek-R1-Distill-Qwen-32B-IQ3_XS.gguf13.7 GB3084ca1a2e1c3172c20e42f64f5bd6ad51276a143c9de3abcb3d537c11a0993e
DeepSeek-R1-Distill-Qwen-32B-IQ4_NL.gguf18.7 GBbb45364875d9e2d8518659aaeadafaadba1574092e0ca9b6d5265d6c47f58a73
DeepSeek-R1-Distill-Qwen-32B-IQ4_XS.gguf17.7 GB91eb0de25c80c0b33a3e89777d96040622f8fa70c71610bf13e19dab297be6b6
DeepSeek-R1-Distill-Qwen-32B-Q2_K.gguf12.3 GB5d3482d2f89bb30c14457ce443441b0c826ac8009d269c4a38a94bd59027126b
DeepSeek-R1-Distill-Qwen-32B-Q2_K_L.gguf13.1 GB3bf816aedf36a9ae0cf4cedfcdefb0dbc43921ea20e332529726868102a1dd10
DeepSeek-R1-Distill-Qwen-32B-Q3_K_L.gguf17.2 GB70b77e6fd9ca78a0fef2480a9052f21358b8a4f0bcea8962403d05b345f2e6c1
DeepSeek-R1-Distill-Qwen-32B-Q3_K_M.gguf15.9 GB4766e528b7e0e16e7304ec228e192a4ee6e776a5b5b5845f00c4576f560aa76a
DeepSeek-R1-Distill-Qwen-32B-Q3_K_S.gguf14.4 GB7f83c6b1f689f6ca8b16088b72e9837921140642243ce1b6f467251e45a28b6a
DeepSeek-R1-Distill-Qwen-32B-Q3_K_XL.gguf17.9 GB5a4fc7e6475446d5d0a43f0a11c4c40d208cee611165a6535286756913486162
DeepSeek-R1-Distill-Qwen-32B-Q4_0.gguf18.7 GB9aa6a3dd22123fb5705ccbc37b2b2c91d8b994f689ed95e1d8dd1bc32f146f43
DeepSeek-R1-Distill-Qwen-32B-Q4_1.gguf20.6 GB80ae4b3680e063bfb173c0afb62a87455d2cdcbdeec92dda4625f0e401dde440
DeepSeek-R1-Distill-Qwen-32B-Q4_K_L.gguf20.4 GB0208dc13febd723a4ad26482fa405cb55cc0b6d68d06b10271e5c737cb5578e9
DeepSeek-R1-Distill-Qwen-32B-Q4_K_M.gguf19.9 GBbed9b0f551f5b95bf9da5888a48f0f87c37ad6b72519c4cbd775f54ac0b9fc62
DeepSeek-R1-Distill-Qwen-32B-Q4_K_S.gguf18.8 GBae5f7e5570239257d554fa1a38ffb8ad634da9dd7e9661e624d1b3bbb6a7b5e8
DeepSeek-R1-Distill-Qwen-32B-Q5_K_L.gguf23.7 GB16e9336945d23b7a86ab6e2bd62d555c623cf44807d95466dfc3ced8a1402135
DeepSeek-R1-Distill-Qwen-32B-Q5_K_M.gguf23.3 GB9f345fb6721d413669afad82154b177cb45e333f49926e38c3586f01c37dfcca
DeepSeek-R1-Distill-Qwen-32B-Q5_K_S.gguf22.6 GBe94e3eeb0ae94955363b32e824ef2d7f1ae8aa973ab71ef6046640dfac7f44d1
DeepSeek-R1-Distill-Qwen-32B-Q6_K.gguf26.9 GB5a7d8d16853575a625fae0812e11fd021618ec254288f5088fb217268b83a1df
DeepSeek-R1-Distill-Qwen-32B-Q6_K_L.gguf27.3 GB2d1dda9ad0affff49d763c2828ffb791299b23c9ced13d0d24587a4381d6cced
DeepSeek-R1-Distill-Qwen-32B-Q8_0.gguf34.8 GB5b321bce035d98e61636e15cd468467e57dfaee4ea28acd42edd7edd1b6098d7
DeepSeek-R1-Distill-Qwen-32B-bf16/DeepSeek-R1-Distill-Qwen-32B-bf16-00001-of-00002.gguf39.9 GB05ecfb6b243833a042bf6e5b24ee359a4b1cee1d15b657dd11e5581fc44e2119
DeepSeek-R1-Distill-Qwen-32B-bf16/DeepSeek-R1-Distill-Qwen-32B-bf16-00002-of-00002.gguf25.7 GB7d7eb11ac21b650a27b55121b1215965109c267defb1dd512ccd6be4d1c477ec
DeepSeek-R1-Distill-Qwen-32B.imatrix15.0 MB715ded400bc723688ebe32515b1e274db4ed99e071f5de4beebb135b459920ca
README.md14.7 KB2f3c131605537200b9997009ce82a962f7e9009e8b3c43d442526b9e4b788eb3
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
1dc8cf9ffa pinned now manifest updated, same commit2026-11-14 21:27 UTC30 files · 519 GB
1dc8cf9ffa2026-11-14 21:27 UTC30 files · 519 GB

Use it

$ fw get bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF
# manifest sha256 29990db1c5b9… · 519 GB · 30 files
$ fw verify ./DeepSeek-R1-Distill-Qwen-32B-GGUF bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF.json --dir ./DeepSeek-R1-Distill-Qwen-32B-GGUF $ python3 fw.py verify ./DeepSeek-R1-Distill-Qwen-32B-GGUF ./bartowski__DeepSeek-R1-Distill-Qwen-32B-GGUF.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF --revision 1dc8cf9ffa5dd333057ea1b09ccf4772d8726dec --local-dir ./DeepSeek-R1-Distill-Qwen-32B-GGUF
llama.cppDownload one GGUF variant (DeepSeek-R1-Distill-Qwen-32B-IQ2_XXS, 9.03 GB), then serve it.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF --include "DeepSeek-R1-Distill-Qwen-32B-IQ2_XXS*" --local-dir ./DeepSeek-R1-Distill-Qwen-32B-GGUF $ llama-server -m ./DeepSeek-R1-Distill-Qwen-32B-GGUF/DeepSeek-R1-Distill-Qwen-32B-IQ2_XXS.gguf
OllamaMake an Ollama model from the downloaded file.
$ printf 'FROM ./DeepSeek-R1-Distill-Qwen-32B-GGUF/DeepSeek-R1-Distill-Qwen-32B-IQ2_XXS.gguf\n' > Modelfile $ ollama create deepseek-r1-distill-qwen-32b-gguf -f Modelfile
Run at home

License

Name
MIT License
Id
MIT
Class
Permissive
Redistribution
yes
Listing
Hosted (signed)
Decided
Inherited from the base model deepseek-ai/DeepSeek-R1-Distill-Qwen-32B: the strictest license along the lineage applies. 2026-11-14
Text in
README.md
Text sha256
2f3c131605537200b9997009ce82a962f7e9009e8b3c43d442526b9e4b788eb3
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: bartowski <huggingface.co/bartowski>Newsgroups: alt.freeweights.deepseekSubject: DeepSeek-R1-Distill-Qwen-32B-GGUFDate: 2026-11-14Commit: 1dc8cf9Signed: 2DC28CC7B26A283B ✓
Upstream
bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF
Pinned commit
1dc8cf9ffa5dd333057ea1b09ccf4772d8726dec
Manifest
DeepSeek-R1-Distill-Qwen-32B-GGUF.json · .minisig
Manifest sha256
29990db1c5b991f7866e4b427ca5f94baf66163e02603da5755cb57471deafc9
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-14 21:27 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF.json $ curl -O https://freeweights.allternit.com/manifests/bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF.json.minisig $ minisign -Vm DeepSeek-R1-Distill-Qwen-32B-GGUF.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-R1-Distill-Qwen-32B · 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 bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. DeepSeek-R1-Distill-Qwen-32B-GGUF.json. Provenance. Takedown history: none.
Source
huggingface.co/bartowski/DeepSeek-R1-Distill-Qwen-32B-GGUF

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