CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.kimi / Kimi-Linear-48B-A3B-Instruct-4bit
From: mlx-community <huggingface.co/mlx-community>Newsgroups: alt.freeweights.kimiSubject:

Kimi-Linear-48B-A3B-Instruct-4bit

Date: 2026-11-15Commit: 919175dSigned: 2DC28CC7B26A283B ✓

Text generation model with 49.1B parameters. Safetensors weights, 27.6 GB in all. Published by mlx-community under MIT.

49.1B params27.6GBKimi-Linear-48B-A3B-Instruct-4bitsafetensorspermissive
FactsParams49.1BSize27.6 GBFormatssafetensorsTasktext generationLicenseMITFrom sourcepermissive
$ fw get mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit
# manifest sha256 5e4168573a5e… · 27.6 GB · 18 files
$ fw verify ./Kimi-Linear-48B-A3B-Instruct-4bit mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit
From sourceHosted (signed)mlx-community/Kimi-Linear-48B-A3B-Instruct-4bitLast check: 2026-11-15 05:12 UTC
Run at homeGet help

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

Quantized of Kimi-Linear-48B-A3B-Instruct.

Works with

Run24Fine-tune29Convert15Evaluate13Scan13Apps58
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
27.6 GB
Weights
27.6 GB (safetensors)
Parameters
49.1B
Formats
safetensors
Task
text generation
License
MIT (permissive)
Files
18, each with a sha256
Publisher
mlx-community
Added
2026-11-15
Section
alt.freeweights.kimi
Lineage

Original release from the publisher.

Base model: moonshotai/Kimi-Linear-48B-A3B-Instruct

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 05:12 UTCpassed

Model card

From the publisher at commit 919175d. Shown as plain text with links; images open at the publisher.

mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit#

This model mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit was converted to MLX format from moonshotai/Kimi-Linear-48B-A3B-Instruct using mlx-lm version 0.28.4.

Use with mlx#

pip install mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/Kimi-Linear-48B-A3B-Instruct-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

18 files · 27.6 GBWeights 27.6 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 18 files with their sha256
FileSizesha256
.gitattributes1.52 KB11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361
README.md939 Be715ebf82e761cbb4282904a75ec977baa647aa08ff552ebf58e5be77acd5fbb
chat_template.jinja1.85 KB75e3cd76bd26b77c02507eb35073b66c8ffd0c88eab185ebc6f8c6e425161809
config.json7.49 KB40ca901b329d5b5d10804c6ce6a68be20f751966ad92a7fb604e9a2b3b02951d
configuration_kimi.py4.95 KB79422aca3ee6c89d201e0c15c4c9a6db517ba83d87ecdc4e41fa0f71297238d9
generation_config.json147 B5d7571b3479fd7263e52c4b2e829c2ae64143abcb0d8d9813258fc4b7994cffe
model-00001-of-00006.safetensors5.15 GBc2e67b2c1b431e78a2331b837bb75b8c0e3cd80566379feb255a3663adb0f1e7
model-00002-of-00006.safetensors5.22 GBd4f05018a6aec2ea12eb3866e3d31ba8477faca1396e18246f9581fffc7649f3
model-00003-of-00006.safetensors5.22 GBf3724036209b2b49d4d9cbdd1d2f5e0702d8626441ac6c343144ee76deee9e2b
model-00004-of-00006.safetensors5.22 GBf5170b87ec6456eec97104487bef4f5129f3ba13157b7c0cdd5c3f06fd3f79bb
model-00005-of-00006.safetensors5.22 GB43428e7d22d9dde1377d0eee2544be9ed3a6a439cf27b2f3e5385b7baff450a1
model-00006-of-00006.safetensors1.60 GBe36748ad3261fc25b54c34e384785d183fe61bd9603197d4e5097af2dd3546ee
model.safetensors.index.json126 KB2e801d6ba423a83d117c2c6c59d24fd543ee3d2da548a22105f2b972a0a5a565
modeling_kimi.py40.8 KBde8607248bb5fdcf1bc46bd5a7afc810bdb4bdadd3fa0198dcbb8598a1c5f9f5
special_tokens_map.json760 B3e1dbdccb7f20d5f7062493dd7e206fd6306d35a62e3e0b33897a697197e34ae
tiktoken.model2.80 MBb6c497a7469b33ced9c38afb1ad6e47f03f5e5dc05f15930799210ec050c5103
tokenization_kimi.py12.5 KBb6b304f63d355f29f5f5aa27694337c824c90e1ac80559ca537909824729761f
tokenizer_config.json3.69 KB5ba77d5822137ac3cb9337d2b84561d10425562cf8b8f2c74c91a832516a778d
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
919175d754 pinned now2026-11-15 05:12 UTC18 files · 27.6 GB

Use it

$ fw get mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit
# manifest sha256 5e4168573a5e… · 27.6 GB · 18 files
$ fw verify ./Kimi-Linear-48B-A3B-Instruct-4bit mlx-community/Kimi-Linear-48B-A3B-Instruct-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/Kimi-Linear-48B-A3B-Instruct-4bit.json --dir ./Kimi-Linear-48B-A3B-Instruct-4bit $ python3 fw.py verify ./Kimi-Linear-48B-A3B-Instruct-4bit ./mlx-community__Kimi-Linear-48B-A3B-Instruct-4bit.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit --revision 919175d75427f1c10cfc17e7254b53295edddf72 --local-dir ./Kimi-Linear-48B-A3B-Instruct-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/Kimi-Linear-48B-A3B-Instruct-4bit", revision="919175d75427f1c10cfc17e7254b53295edddf72") model = AutoModelForCausalLM.from_pretrained("mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit", revision="919175d75427f1c10cfc17e7254b53295edddf72", device_map="auto")
vLLMServe an OpenAI-compatible API.
$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit --revision 919175d75427f1c10cfc17e7254b53295edddf72
Run at home

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
e715ebf82e761cbb4282904a75ec977baa647aa08ff552ebf58e5be77acd5fbb
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.kimiSubject: Kimi-Linear-48B-A3B-Instruct-4bitDate: 2026-11-15Commit: 919175dSigned: 2DC28CC7B26A283B ✓
Upstream
mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit
Pinned commit
919175d75427f1c10cfc17e7254b53295edddf72
Manifest
Kimi-Linear-48B-A3B-Instruct-4bit.json · .minisig
Manifest sha256
5e4168573a5e87c9951fe89fc7f2618677f612842c0d04d64963a5f5505e37e5
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 05:12 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit.json $ curl -O https://freeweights.allternit.com/manifests/mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit.json.minisig $ minisign -Vm Kimi-Linear-48B-A3B-Instruct-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

moonshotai/Kimi-Linear-48B-A3B-Instruct · 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/Kimi-Linear-48B-A3B-Instruct-4bit
Open in Freeweights Studio

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

Signed manifest
minisign, key 2DC28CC7B26A283B. Kimi-Linear-48B-A3B-Instruct-4bit.json. Provenance. Takedown history: none.
Source
huggingface.co/mlx-community/Kimi-Linear-48B-A3B-Instruct-4bit

Related models

All of alt.freeweights.kimi →
From: laion <huggingface.co/laion>Subject: kimi-k2-r2egym_sandboxes-maxeps-32k
From: laion <huggingface.co/laion>Subject: kimi-k2t-freelancer-32ep-32k
From: laion <huggingface.co/laion>Subject: Kimi-K2T-ling-coder-sft-sandboxes-1-maxeps-32k
From: laion <huggingface.co/laion>Subject: Kimi-K2T-neulab-agenttuning-kg-sandboxes-maxeps-32k
From: laion <huggingface.co/laion>Subject: Kimi-K2T-neulab-agenttuning-mind2web-sandboxes-maxeps-32k
From: laion <huggingface.co/laion>Subject: Kimi-K2T-neulab-agenttuning-webshop-sandboxes-maxeps-32k