CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.mimo / XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF
From: bartowski <huggingface.co/bartowski>Newsgroups: alt.freeweights.mimoSubject:

XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF

Date: 2026-11-15Commit: dc67b07Signed: 2DC28CC7B26A283B ✓

Image text to text model. Gguf weights, 124 GB in all. Published by bartowski under MIT.

mimo124GBXiaomiMiMo_MiMo-VL-7B-RL-2508-GGUFggufpermissive
FactsParamsnot statedSize124 GBFormatsggufTaskimage text to textLicenseMITFrom sourcepermissive
$ fw get bartowski/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF
# manifest sha256 15ff5b1b062e… · 124 GB · 29 files
$ fw verify ./XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF bartowski/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF
From sourceHosted (signed)bartowski/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUFLast check: 2026-11-15 04:56 UTC
Run at homeGet help

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

Quantized of MiMo-VL-7B-RL-2508.

Works with

Run13Convert1Evaluate13Scan12Apps62

Convert with

ggml
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
124 GB
Weights
5.16 MB (smallest of 25 GGUF variants)
Parameters
not stated
Formats
gguf
Task
image text to text
License
MIT (permissive)
Files
29, each with a sha256
Publisher
bartowski
Added
2026-11-15
Section
alt.freeweights.mimo
Lineage

Original release from the publisher.

Base model: XiaomiMiMo/MiMo-VL-7B-RL-2508

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 04:56 UTCpassed

Model card

From the publisher at commit dc67b07. Shown as plain text with links; images open at the publisher.

Llamacpp imatrix Quantizations of MiMo-VL-7B-RL-2508 by XiaomiMiMo#

Using llama.cpp release b6317 for quantization.

Original model: https://huggingface.co/XiaomiMiMo/MiMo-VL-7B-RL-2508

All quants made using imatrix option with dataset from here combined with a subset of combined_all_small.parquet from Ed Addario here

Run them in LM Studio

Run them directly with llama.cpp, or any other llama.cpp based project

Prompt format#

No prompt format found, check original model page

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

FilenameQuant typeFile SizeSplitDescription
MiMo-VL-7B-RL-2508-bf16.ggufbf1615.25GBfalseFull BF16 weights.
MiMo-VL-7B-RL-2508-Q8_0.ggufQ8_08.11GBfalseExtremely high quality, generally unneeded but max available quant.
MiMo-VL-7B-RL-2508-Q6_K_L.ggufQ6_K_L6.56GBfalseUses Q8_0 for embed and output weights. Very high quality, near perfect, recommended.
MiMo-VL-7B-RL-2508-Q6_K.ggufQ6_K6.26GBfalseVery high quality, near perfect, recommended.
MiMo-VL-7B-RL-2508-Q5_K_L.ggufQ5_K_L5.83GBfalseUses Q8_0 for embed and output weights. High quality, recommended.
MiMo-VL-7B-RL-2508-Q5_K_M.ggufQ5_K_M5.45GBfalseHigh quality, recommended.
MiMo-VL-7B-RL-2508-Q5_K_S.ggufQ5_K_S5.33GBfalseHigh quality, recommended.
MiMo-VL-7B-RL-2508-Q4_K_L.ggufQ4_K_L5.15GBfalseUses Q8_0 for embed and output weights. Good quality, recommended.
MiMo-VL-7B-RL-2508-Q4_1.ggufQ4_14.89GBfalseLegacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon.
MiMo-VL-7B-RL-2508-Q4_K_M.ggufQ4_K_M4.68GBfalseGood quality, default size for most use cases, recommended.
MiMo-VL-7B-RL-2508-Q3_K_XL.ggufQ3_K_XL4.68GBfalseUses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability.
MiMo-VL-7B-RL-2508-Q4_K_S.ggufQ4_K_S4.48GBfalseSlightly lower quality with more space savings, recommended.
MiMo-VL-7B-RL-2508-Q4_0.ggufQ4_04.47GBfalseLegacy format, offers online repacking for ARM and AVX CPU inference.
MiMo-VL-7B-RL-2508-IQ4_NL.ggufIQ4_NL4.47GBfalseSimilar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference.
MiMo-VL-7B-RL-2508-IQ4_XS.ggufIQ4_XS4.26GBfalseDecent quality, smaller than Q4_K_S with similar performance, recommended.
MiMo-VL-7B-RL-2508-Q3_K_L.ggufQ3_K_L4.14GBfalseLower quality but usable, good for low RAM availability.
MiMo-VL-7B-RL-2508-Q3_K_M.ggufQ3_K_M3.85GBfalseLow quality.
MiMo-VL-7B-RL-2508-Q2_K_L.ggufQ2_K_L3.68GBfalseUses Q8_0 for embed and output weights. Very low quality but surprisingly usable.
MiMo-VL-7B-RL-2508-IQ3_M.ggufIQ3_M3.65GBfalseMedium-low quality, new method with decent performance comparable to Q3_K_M.
MiMo-VL-7B-RL-2508-Q3_K_S.ggufQ3_K_S3.53GBfalseLow quality, not recommended.
MiMo-VL-7B-RL-2508-IQ3_XS.ggufIQ3_XS3.40GBfalseLower quality, new method with decent performance, slightly better than Q3_K_S.
MiMo-VL-7B-RL-2508-IQ3_XXS.ggufIQ3_XXS3.15GBfalseLower quality, new method with decent performance, comparable to Q3 quants.
MiMo-VL-7B-RL-2508-Q2_K.ggufQ2_K3.08GBfalseVery low quality but surprisingly usable.
MiMo-VL-7B-RL-2508-IQ2_M.ggufIQ2_M2.87GBfalseRelatively low quality, uses SOTA techniques to be surprisingly 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/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF --include "XiaomiMiMo_MiMo-VL-7B-RL-2508-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/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF --include "XiaomiMiMo_MiMo-VL-7B-RL-2508-Q8_0/*" --local-dir ./

You can either specify a new local-dir (XiaomiMiMo_MiMo-VL-7B-RL-2508-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, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.

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.

Thank you to LM Studio for sponsoring my work.

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

Files

29 files · 124 GBWeights 5.16 MB (smallest of 25 GGUF variants). Sizes and sha256 from the signed manifest.
Show all 29 files with their sha256
FileSizesha256
.gitattributes3.62 KB6ec84b954c8332c11e09897204b742cb9da5fea784e15eee17f20259afad3eec
README.md14.1 KBf420e78b1d3c377f784bf1c6ccc65e5cebdeb1c505921fda29d7e53aeee93921
XiaomiMiMo_MiMo-VL-7B-RL-2508-IQ2_M.gguf2.87 GB748484dae5aa372e2117766e7f58bce749492d855643c421b76e4d09226030ef
XiaomiMiMo_MiMo-VL-7B-RL-2508-IQ3_M.gguf3.65 GB7f51116160680f94385aa6d6ede6a1d533664ff49fb4bb98ed4867ea1dc23517
XiaomiMiMo_MiMo-VL-7B-RL-2508-IQ3_XS.gguf3.40 GB7800f454ca1037c7e2fa956204bf658888578bb549d5dbecfb9c60ccd62eea3c
XiaomiMiMo_MiMo-VL-7B-RL-2508-IQ3_XXS.gguf3.15 GB5c82915554387512f2c2fc1e2400f17148b2b3fcae2384649fa705d5e46e4359
XiaomiMiMo_MiMo-VL-7B-RL-2508-IQ4_NL.gguf4.47 GB2441a62231c7e61f5f9d34cb6a112c52a200b4752833354a5d1e52c3a7732d82
XiaomiMiMo_MiMo-VL-7B-RL-2508-IQ4_XS.gguf4.26 GB42072a604618a3fb8ca0f33d5fa9eb7fc265d87d4b55ee6258b7e7f0fe7b1314
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q2_K.gguf3.08 GB11b22b152a93da58d211141ae44243269bd72c5831ae7c9d3fa2bd8f7d66fe78
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q2_K_L.gguf3.68 GBcb10226a9f5accb80593eb727aa1ac4241d05757536d7b5bd09388df58750bd6
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q3_K_L.gguf4.14 GB25305d28eee7121874d07fec4e4c23d48d705f0cb2d2bad9fe82e7d1b7ec99e1
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q3_K_M.gguf3.85 GBebd92b211c9d31bd28fbae3df38b59a0b138bafcc3bffa3b6a0d67dcf2a2893e
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q3_K_S.gguf3.53 GB9e4b35df9d069db9fb1669fdb66338e74c2e1ffbc5f44584e1439ea1e8b54f77
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q3_K_XL.gguf4.68 GBb01b0b2a568962a61828d75713d899812f06372eb67a1558761d6e29514c6d5c
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q4_0.gguf4.47 GBb93f4b2ded0e023ba42458d667fb84c643d0818a3cf63d6d719fecd272480609
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q4_1.gguf4.89 GB71bcda6d3b8fe9583e85fcaa53ec20c8c146f450a27b2f0f1ca3f31b5724bacb
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q4_K_L.gguf5.15 GBa3afdb77dc27f4142cecb30c13b23264f06c7e4ebd4e1904a224a7cc5d7a0412
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q4_K_M.gguf4.68 GB78f21f8dd95a3ab3b617b791459525438157ad5acfd34c5ac94958326f93235b
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q4_K_S.gguf4.48 GB212f80050cf8c179f62427041ba9730fd9bcc62a44d711ce29c2beca77d936f6
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q5_K_L.gguf5.83 GBd277267a478ca230d28c79a7dc9dc52324486e6967afb75342c73fbd0beac847
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q5_K_M.gguf5.45 GBed7015b81cbc6bff8abdfaf898258e44318f36be24886b3306921b3871d48089
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q5_K_S.gguf5.33 GB400d8b42fba0f098058370559f3918c23c5c6eb88c2c814ed19147c56d34b3e6
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q6_K.gguf6.26 GBeabe1b58e7ae1dfb85537e36261a96c062b2d8ce38f77caddf700891fa98b865
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q6_K_L.gguf6.56 GB7118789f18c48e82c0267e04893427386dad61b9f2d85d4f578ef71b4a116c22
XiaomiMiMo_MiMo-VL-7B-RL-2508-Q8_0.gguf8.11 GB612db7eb8f0a6805b782346861ecbe7bf84b61c8f0b57c8a5290fe661dcf066d
XiaomiMiMo_MiMo-VL-7B-RL-2508-bf16.gguf15.3 GB7ced6491884801ee706eb4c1da54901fdabfad23816f64370ce73787dfb6b6ac
XiaomiMiMo_MiMo-VL-7B-RL-2508-imatrix.gguf5.16 MB72254a7922371aa4570992e70a6ad7261677880e70d432bb707a1419d0c1798d
mmproj-XiaomiMiMo_MiMo-VL-7B-RL-2508-bf16.gguf1.37 GB9ef747c6fad49af289cae4da97ddecbff7b2731e66563cb71ab8dbe1a8150a1c
mmproj-XiaomiMiMo_MiMo-VL-7B-RL-2508-f16.gguf1.37 GB68f983965002dfc3819125f0d4a61882fcc8de60f17fc05f9b8715f46645a66b
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
dc67b07909 pinned now2026-11-15 04:56 UTC29 files · 124 GB

Use it

$ fw get bartowski/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF
# manifest sha256 15ff5b1b062e… · 124 GB · 29 files
$ fw verify ./XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF bartowski/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/bartowski/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF.json --dir ./XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF $ python3 fw.py verify ./XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF ./bartowski__XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download bartowski/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF --revision dc67b07909a98a50d5798570f50695a38440c0b5 --local-dir ./XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF
llama.cppDownload one GGUF variant (XiaomiMiMo_MiMo-VL-7B-RL-2508-imatrix, 5.16 MB), then serve it.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download bartowski/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF --include "XiaomiMiMo_MiMo-VL-7B-RL-2508-imatrix*" --local-dir ./XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF $ llama-server -m ./XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF/XiaomiMiMo_MiMo-VL-7B-RL-2508-imatrix.gguf
OllamaMake an Ollama model from the downloaded file.
$ printf 'FROM ./XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF/XiaomiMiMo_MiMo-VL-7B-RL-2508-imatrix.gguf\n' > Modelfile $ ollama create xiaomimimo-mimo-vl-7b-rl-2508-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 XiaomiMiMo/MiMo-VL-7B-RL-2508: the strictest license along the lineage applies. 2026-11-15
Text in
README.md
Text sha256
f420e78b1d3c377f784bf1c6ccc65e5cebdeb1c505921fda29d7e53aeee93921
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.mimoSubject: XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUFDate: 2026-11-15Commit: dc67b07Signed: 2DC28CC7B26A283B ✓
Upstream
bartowski/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF
Pinned commit
dc67b07909a98a50d5798570f50695a38440c0b5
Manifest
XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF.json · .minisig
Manifest sha256
15ff5b1b062e0758b4ec869c352a5b41175e7823fcc6b0b6dfc5ea37df4714ff
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 04:56 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/bartowski/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF.json $ curl -O https://freeweights.allternit.com/manifests/bartowski/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF.json.minisig $ minisign -Vm XiaomiMiMo_MiMo-VL-7B-RL-2508-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

XiaomiMiMo/MiMo-VL-7B-RL-2508 · 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/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF.json. Provenance. Takedown history: none.
Source
huggingface.co/bartowski/XiaomiMiMo_MiMo-VL-7B-RL-2508-GGUF

Related models

All of alt.freeweights.mimo →
From: XiaomiMiMo <huggingface.co/XiaomiMiMo>Subject: MiMo-7B-Base
From: XiaomiMiMo <huggingface.co/XiaomiMiMo>Subject: MiMo-V2-Flash
From: unsloth <huggingface.co/unsloth>Subject: MiMo-V2-Flash-GGUF
From: XiaomiMiMo <huggingface.co/XiaomiMiMo>Subject: MiMo-V2.5
From: XiaomiMiMo <huggingface.co/XiaomiMiMo>Subject: MiMo-V2.5-ASR
From: mlx-community <huggingface.co/mlx-community>Subject: MiMo-V2.5-ASR-MLX-8bit