CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.glm / GLM-5.3-Flash-4bit
From: mlx-community <huggingface.co/mlx-community>Newsgroups: alt.freeweights.glmSubject:

GLM-5.3-Flash-4bit

Date: 2026-11-15Commit: fa04cbcSigned: 2DC28CC7B26A283B ✓

Image text to text model with 347.9B parameters. Safetensors weights, 204 GB in all. Published by mlx-community under MIT.

347.9B params204GBGLM-5.3-Flash-4bitsafetensorspermissive
FactsParams347.9BSize204 GBFormatssafetensorsTaskimage text to textLicenseMITFrom sourcepermissive
$ fw get mlx-community/GLM-5.3-Flash-4bit
# manifest sha256 1da72414f9ac… · 204 GB · 73 files
$ fw verify ./GLM-5.3-Flash-4bit mlx-community/GLM-5.3-Flash-4bit
From sourceHosted (signed)mlx-community/GLM-5.3-Flash-4bitLast check: 2026-11-15 04:56 UTC
Run with Allternit CloudChat with itGet help

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

This model is served on Allternit Cloud as glm-5.3-flash (the original model it was made from). Get a key in the console, then call the OpenAI-compatible API.

$ curl https://api.allternit.com/v1/chat/completions \ -H "Authorization: Bearer $ALLTERNIT_API_KEY" \ -H "Content-Type: application/json" \ -d '{"model": "glm-5.3-flash", "messages": [{"role": "user", "content": "Hello"}]}'

Quantized of GLM-5.3-Flash.

Works with

Run23Fine-tune29Convert15Evaluate13Scan13Apps62
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
204 GB
Weights
204 GB (safetensors)
Parameters
347.9B
Formats
safetensors
Task
image text to text
License
MIT (permissive)
Files
73, each with a sha256
Publisher
mlx-community
Added
2026-11-15
Section
alt.freeweights.glm
Lineage

Original release from the publisher.

Base model: zai-org/GLM-5.3-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 04:56 UTCpassed

Model card

From the publisher at commit fa04cbc. Shown as plain text with links; images open at the publisher.

mlx-community/GLM-5.3-Flash-4bit#

zai-org/GLM-5.3-Flash in MLX affine format, mixed 4/5/6-bit, group size 64.

The weights are the 4-bit variant of orcarouter/GLM-5.3-Flash-MLX (revision c80f6810b1a95b5be9042761becc6aa78d189782, the 4-bit/ build, which is also the repo root there). The quantisation is orcarouter's. The 62 safetensors shards are byte-identical to that build (sha256 checked).

Change from the source: config.json now lists the per-module widths of the MTP layer (layer 45) in quantization and quantization_config. The source config named them for layers 3-44 only. The tensors are not changed.

Size: 62 shards, 203,992,076,296 bytes (190 GiB). It needs a Mac with 256 GB of memory or more.

Widths#

Read from each tensor's shapes (bits = 32 * weight_columns / (group_size * scales_columns)), group size 64 throughout.

ModulesLayersBits
Routed experts gate_proj, up_proj3-454
Routed experts down_proj3-455
Shared expert gate_proj, up_proj, down_proj3-456
Dense MLP gate_proj, up_proj0-24
Dense MLP down_proj0-25
Sparse attention q_a_proj, q_b_proj, kv_a_proj_with_mqa, o_proj3, 7, 11, ..., 43, 45 (12 layers)4
Linear attention layers, kv_b_proj, indexer, router, mHC, norms, embed_tokens, lm_head, MTP eh_proj, vision towerallbf16

Layer 45 is the MTP (next-token prediction) layer.

Usage#

mlx-lm does not support the glm5_next architecture. mlx-vlm has a glm5_next model; see the orcarouter model card for its use.

License#

MIT, the same as the base model.

Files

73 files · 204 GBWeights 204 GB (safetensors). Sizes and sha256 from the signed manifest.
Show all 73 files with their sha256
FileSizesha256
.gitattributes1.64 KBfe81d528e7ee055bbcf6a310afe0f2808423a4a16d45225f29c285926abef08d
README.md1.93 KB4fabe39658e33cff715ab572d13af8b4dc94b0524ba23b18a46c502b847bd5ee
chat_template.jinja8.62 KB41cff9af7b3a86c96751b107a8444f245fbda0bd5320b636a5bb1f7f4ba1a5c3
config.json40.7 KBff382668c5c53c774c718ed662bf336b08a41c01f515bde9a8a6c7948377f2eb
generation_config.json194 B230c30609ecbbb9e6583bedde8e7bdda0c6eb8fe5fad0eaeb3d1b293d751cb4f
model-00001-of-00062.safetensors4.27 GB40e146ab2951b8bf550094f2a0b8f4684348b2122a7c5fa9cc2ccff56c6e4497
model-00002-of-00062.safetensors3.37 GBf31c1467eb18505ecb7c1cbdf1ebdcbef2e2397d669d7e4b5d41d116832f3d6e
model-00003-of-00062.safetensors3.32 GBd4025e891d4754c0b63dec2c23672abe0ac6fe29da3adb623de42b121c75cfd5
model-00004-of-00062.safetensors3.36 GBf8f1032ce19bf338e8135bcaa7cab466988156d8151751e948e13ed934cad73f
model-00005-of-00062.safetensors3.26 GBbad704b7044fe06bc9e8b0d3db8605f39f2f47eb94bf9c8a8c28c42b6fca6985
model-00006-of-00062.safetensors3.24 GBc10f2d94a9143d3ed6b2e158f6718076995adb94d1eb0885b313d8b5a2e76d93
model-00007-of-00062.safetensors3.36 GB781e26f58cd93647f4cbd9a981d0ecd984bec0ccae10b659efef978d5246895e
model-00008-of-00062.safetensors3.36 GB8bbe3cd90f8a4b37a9ac5660c5e04e0af84b96ddc36608980d5d4c09454db492
model-00009-of-00062.safetensors3.24 GB2235c379ac9f26d269950198e3079e444b6e42f163ab6f50a76497e4092a1189
model-00010-of-00062.safetensors3.35 GBf0fb41f5f53639d289f13614b7ca46841c62f9a8f16c34e67db877f2b6a15c09
model-00011-of-00062.safetensors3.26 GB5cd42a40d364167780b4c050a3bae3cf643160ad9f2f6d9ae9053103e0c6baf7
model-00012-of-00062.safetensors3.36 GB6851010855db522ed0207f442f1e93d41cac9fae6473f1b9a3d53367348d3d9c
model-00013-of-00062.safetensors3.24 GBa4e188580c02b79b53f34eb578522b04122f97a4ede63a0ba84df37889ffc0f8
model-00014-of-00062.safetensors3.36 GBe7cb62eec0388a239a3a96dbcbfbdf87ef0295491afd6e123cadc899f94d92e3
model-00015-of-00062.safetensors3.36 GB88fe9a238be6fe00003c1a4cc1e9d329743a51244ad87eaba1cb8eb0c3e430e2
model-00016-of-00062.safetensors3.24 GB743815f49c1cfb59652742fc098c70291ad0194e3daca70c81a60214949a311b
model-00017-of-00062.safetensors3.37 GBa50583019066f2bc9424377fdeb97c52280af312e3761bea23224d7e2c154330
model-00018-of-00062.safetensors3.36 GB138fd451b0265c4c97c0c1b0930a7c720c98020d87b95579c2f5dc1d64b76fe8
model-00019-of-00062.safetensors3.36 GBd5933d1a142392deedc93640323b86e171b2eed237d8e163e9773a534ae6cec4
model-00020-of-00062.safetensors3.24 GB6b50ac3c4abf840c9cc2405defc3912c5e7a07e791659700621c5e342f258730
model-00021-of-00062.safetensors3.36 GBc44253a8b949b2d656834c44170f400d6c11f837065d88fe251b9dc26d62d957
model-00022-of-00062.safetensors3.26 GB22c5054b41d8fe7efc09edbe0ff949e888602dbc79067a8484f140a04969070d
model-00023-of-00062.safetensors3.24 GB8f3fa5da0c7f5a204186c1127c1b39af2b2fe1dc246852e4c2e1ed63b9e8b56e
model-00024-of-00062.safetensors3.36 GBd844188d8e888e08f386fbb4ab379324a64d9d28f3ea935c28be19be0380fbdb
model-00025-of-00062.safetensors3.36 GB0b6f0e316ff9b31e88bd2e9b50b2e39616926dce7b3b943a948ec79dbbad51b9
model-00026-of-00062.safetensors3.36 GBa327f689c7f8f7dc9edcfba8bfc5e644cab40699a193c590aef4c62bba72b31c
model-00027-of-00062.safetensors3.24 GBb60ae28d3188f3bc8daf46524d48c83730c8e42472b5195bbf450d7bd1661365
model-00028-of-00062.safetensors3.26 GBcd377ef04aa09aeac0ef1907386df64ea1f896c90870f7ebba4ba0080abd95b5
model-00029-of-00062.safetensors3.36 GB034fe28869fe0e7334dbc10acb212eb2751eea7cac39f735881d15ef85997ed3
model-00030-of-00062.safetensors3.24 GBe281f7512a23ac0553c51663412502c470dc7cff08cab2c02b4e4a5c6952b1e5
model-00031-of-00062.safetensors3.36 GB2e5de9d6d4e62dd149864f4e39029d40474f42673c567239cadbdf799c8ca14c
model-00032-of-00062.safetensors3.26 GB31ab6a03a1e94fc86c8d6c20a619f334baa606c33d4506a6f016356b2095f0bc
model-00033-of-00062.safetensors3.36 GBb0fc013479f13b7eded07fa3c55c13581e26b2709f1e4845137aa5445bbdb313
model-00034-of-00062.safetensors3.24 GBf8aa0d64d0c2283950769d4a5140e828dea66ae7a0645138384628a6482d5383
model-00035-of-00062.safetensors3.26 GBc29db5aa63457f470343e452d21e6a058b30e202e81f746d66b2c7e42e969501
model-00036-of-00062.safetensors3.36 GBc2c1a04d589cab3305a635d8eccc4445e1d2b52ffb07a9659b66de6e9d8bad14
model-00037-of-00062.safetensors3.24 GBd636e2158c98e505a881c5d6071b9f99dee409843d216849f1226efdb418677c
model-00038-of-00062.safetensors3.36 GB7cc0312293990f7a274a5c6c64ac3e48541a38e30e7740b3f565179a19f5acf7
model-00039-of-00062.safetensors3.36 GB489ce0e8f86e81b3da66101d108e0ffebe30e0587ed359998420d6b71c66877b
model-00040-of-00062.safetensors3.26 GB6140e807ccd642b694c25f8f4485a7e13abb3e3bd7e2230bfba5ad8aa62b3b07
model-00041-of-00062.safetensors3.24 GBe0b46d196fae106afa9fa2454e896758f9ca95f5b1c8c0ae2b52ba412cd1d397
model-00042-of-00062.safetensors3.36 GB9b3319c95931308a7dec14281afae78cfdaa7db00b2b29eae17e3a52fd63bb63
model-00043-of-00062.safetensors3.36 GBe5f1a5d6b905ad543402127fe672de729b0e21850e11fda870d6fb6708d0f8bd
model-00044-of-00062.safetensors3.24 GBeff8baf86414029418ca16862b23b7f512a1eae4d5e567c98f7a19b3ba50a6e4
model-00045-of-00062.safetensors3.36 GBdd0d130364e5c257c8c9774c2047223d2436eba66b7afe2a64041047f9c21887
model-00046-of-00062.safetensors3.26 GB364025a003f2d84c20083bd23e4a71b737262634e3707fa0dc5d6c713f964ca0
model-00047-of-00062.safetensors3.36 GB962788b2868ad7785cfd93fe902fca731b185169fdf4c6f557dd8d30dacb7b98
model-00048-of-00062.safetensors3.24 GB11443936a53bcc393080eb875b00911270d18629ed44c3e286feca154804a20e
model-00049-of-00062.safetensors3.36 GB6db088ef1f7aeca30ba0ff71c054b2827ac3aea4f642d5e8af15404b011f7d49
model-00050-of-00062.safetensors3.36 GB0e245fa85068ca0b1f8decd4ebc76d3b760f20de17d7b625e1ab2b4b0c08075e
model-00051-of-00062.safetensors3.24 GB1bbae915848ec95451d0d0d7b21f678da9062c4aea74861ba7d33caab047f114
model-00052-of-00062.safetensors3.36 GBaed092f60d5369559d32b235f6b206296484f68076ee41ed7529e616ac87520b
model-00053-of-00062.safetensors3.26 GB9487ba6fee2043f356d151c00b964b78ff708a517f4e8751846dd7700cc83015
model-00054-of-00062.safetensors3.36 GBbff299dbe5a0d922bd17fadd22b18f04f9ba15754b7eab1b8c3c9a0517d08517
model-00055-of-00062.safetensors3.24 GB83663300b4186393e26e4e75f03a6948cd1f5e8746488ad84fb79f83e566f0c3
model-00056-of-00062.safetensors3.36 GB4db4d827af705cde2b1dc57fef8b14c0b66d9042d7a8f2978cae39881b848d08
model-00057-of-00062.safetensors3.36 GB6c63b9ac733c81e7664bbfc2044039a4af518a8ceb0de1e609a0518e75dad2a6
model-00058-of-00062.safetensors3.24 GB11a824512bb345d559a14b324407f90e98557adb6a4d9bf76be1eee90a6ae337
model-00059-of-00062.safetensors3.26 GB27acd6d5c3377c76759ad5ca18ce444a791368f8477c4fa78858a3ed9879a204
model-00060-of-00062.safetensors3.36 GB676bdddecc1d5b2c8d837c40be1cd5151cbe632382bada4ca0b468e04f78c945
model-00061-of-00062.safetensors3.27 GB4eecab41dca94911b6743e5cd56406e30359bd6008785a8880c4bb15ec203a23
model-00062-of-00062.safetensors1.26 GB1027a816aeaaadffbe4ce7b78f30072d5eaeafbc28068e39b714021f5ea1b915
model.safetensors.index.json11.5 MBcacc0f91d635087c63adf6136e51935e251f12b810fbba2f77551c7aaececa1a
preprocessor_config.json573 B0b686d58d19d92e4363983a2558cb15495bfc51536322de17e443114c4e80ed6
processor_config.json909 Baae38374c94b08cc9b0547c6e64f05b951bd9735cea571c6988f5ed552bed3ed
tokenizer.json20.2 MB19e773648cb4e65de8660ea6365e10acca112d42a854923df93db4a6f333a82d
tokenizer_config.json761 B98b1271574f41abf89427ae2dda030d94dc9478f0edc5a8bd240db213c6fd5fc
video_preprocessor_config.json699 Ba7b3c6f410c100b30c29472a3f2ba7733a6abafd0120c011a8a45ebc4fe3e2a3
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
fa04cbc362 pinned now2026-11-15 04:56 UTC73 files · 204 GB

Use it

$ fw get mlx-community/GLM-5.3-Flash-4bit
# manifest sha256 1da72414f9ac… · 204 GB · 73 files
$ fw verify ./GLM-5.3-Flash-4bit mlx-community/GLM-5.3-Flash-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/GLM-5.3-Flash-4bit.json --dir ./GLM-5.3-Flash-4bit $ python3 fw.py verify ./GLM-5.3-Flash-4bit ./mlx-community__GLM-5.3-Flash-4bit.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download mlx-community/GLM-5.3-Flash-4bit --revision fa04cbc3622b16aa7475404060b05e506d81bd4e --local-dir ./GLM-5.3-Flash-4bit
transformersLoad the weights in Python.
import os os.environ["HF_ENDPOINT"] = "https://freeweights.allternit.com" from transformers import AutoModelForImageTextToText, AutoProcessor processor = AutoProcessor.from_pretrained("mlx-community/GLM-5.3-Flash-4bit", revision="fa04cbc3622b16aa7475404060b05e506d81bd4e") model = AutoModelForImageTextToText.from_pretrained("mlx-community/GLM-5.3-Flash-4bit", revision="fa04cbc3622b16aa7475404060b05e506d81bd4e", device_map="auto")
vLLMServe an OpenAI-compatible API.
$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve mlx-community/GLM-5.3-Flash-4bit --revision fa04cbc3622b16aa7475404060b05e506d81bd4e

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
4fabe39658e33cff715ab572d13af8b4dc94b0524ba23b18a46c502b847bd5ee
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.glmSubject: GLM-5.3-Flash-4bitDate: 2026-11-15Commit: fa04cbcSigned: 2DC28CC7B26A283B ✓
Upstream
mlx-community/GLM-5.3-Flash-4bit
Pinned commit
fa04cbc3622b16aa7475404060b05e506d81bd4e
Manifest
GLM-5.3-Flash-4bit.json · .minisig
Manifest sha256
1da72414f9acbb7a2d38502ec0872bb1ebabd1b9b529e8c8e0b4a03d95a4b4ce
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/mlx-community/GLM-5.3-Flash-4bit.json $ curl -O https://freeweights.allternit.com/manifests/mlx-community/GLM-5.3-Flash-4bit.json.minisig $ minisign -Vm GLM-5.3-Flash-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

zai-org/GLM-5.3-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/GLM-5.3-Flash-4bit
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. GLM-5.3-Flash-4bit.json. Provenance. Takedown history: none.
Source
huggingface.co/mlx-community/GLM-5.3-Flash-4bit

Related models

All of alt.freeweights.glm →
From: bartowski <huggingface.co/bartowski>Subject: arcee-ai_GLM-4-32B-Base-32K-GGUF
From: ggml-org <huggingface.co/ggml-org>Subject: AutoGLM-Phone-9B-GGUF
From: zai-org <huggingface.co/zai-org>Subject: cogagent-chat-hf
From: zai-org <huggingface.co/zai-org>Subject: CogView3-Plus-3B
From: zai-org <huggingface.co/zai-org>Subject: cogvlm-base-490-hf
From: laion <huggingface.co/laion>Subject: dev_set_part1_10k_glm_4_7_traces_jupiter