From: <huggingface.co/mlx-community> Newsgroups: alt.freeweights.textSubject: Nanbeige4.1-3B-8bit Date: 2026-11-15Commit: 699b0e7Signed: 2DC28CC7B26A283B ✓ Text generation model with 3.9B parameters. Safetensors weights, 4.20 GB in all. Published by mlx-community under Apache-2.0.
3.9B params 4.20 GB Nanbeige4.1-3B-8bit safetensors permissive Facts Params 3.9B Size 4.20 GB Formats safetensors Task text generation License Apache-2.0 permissive
fw
$ fw get mlx-community/Nanbeige4.1-3B-8bit
# manifest sha256 f54489e9aecc… · 4.20 GB · 9 files
$ fw verify ./Nanbeige4.1-3B-8bit mlx-community/Nanbeige4.1-3B-8bitHosted (signed) mlx-community/Nanbeige4.1-3B-8bit Last check: 2026-11-15 05:04 UTC
Get help: custom fine-tune or private deploy. We review your setup before we quote.
Quantized of Nanbeige4.1-3B .
Works with
Runs with Ranked by fit to this model (safetensors), then by GitHub stars.
Show all 24 Fine-tune with Show all 29
Overview
Size 4.20 GB
Weights 4.18 GB (safetensors)
Parameters 3.9B
Formats safetensors
Task text generation
License Apache-2.0 (permissive)
Files 9, each with a sha256
Publisher mlx-community
Added 2026-11-15
Section alt.freeweights.text Lineage
Original release from the publisher.
Base model: Nanbeige/Nanbeige4.1-3B
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
Badge Host Last check Result huggingface.co 2026-11-15 05:04 UTC passed
Model card From the publisher at commit
699b0e7 . Shown as plain text with links; images open at the publisher.
This model mlx-community/Nanbeige4.1-3B-8bit was converted to MLX format from Nanbeige/Nanbeige4.1-3B using mlx-lm version 0.30.7 .
Use with mlx#
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/Nanbeige4.1-3B-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 9 files · 4.20 GB Weights 4.18 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 9 files with their sha256 File Size sha256 .gitattributes 1.57 KB 34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930 README.md 899 B d9730bb8db566f224a41e93a666471133a84f5ffbcb54e0c3905ae05da06db2a chat_template.jinja 5.53 KB 757f108242fb797d23b79f72c088cb5b4bc37fb3744e29dc8771c1dc2b1cb335 config.json 1.05 KB 5b3ae9cd2a94463ac0bf76d654b4f8d03e758c3cbb67068c1eb6250152c72dc2 generation_config.json 142 B 762de61aa816a028146e845d59b435e11295fbf18e9a7696c0c87df5c85be359 model.safetensors 4.18 GB c02389ae17426b03d1336efb993bfe75868a1aec86411e8bd0eb82f1fa2ff93d model.safetensors.index.json 52.4 KB aa6b7008f3226254cc86891c1685246080579a5c796a4c29c6870c50751b15dc tokenizer.json 18.5 MB 1d8f0326910136aca20831249220b38ce5299527647bc8c6b65404485c479740 tokenizer_config.json 575 B 0e9b61b839c76099029e5426a5ba72dd702116a42a780636393106afdb1a3751
Versions Each time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commit Date Contents 699b0e755a pinned now 2026-11-15 05:04 UTC 9 files · 4.20 GB
Use it fw
$ fw get mlx-community/Nanbeige4.1-3B-8bit
# manifest sha256 f54489e9aecc… · 4.20 GB · 9 files
$ fw verify ./Nanbeige4.1-3B-8bit mlx-community/Nanbeige4.1-3B-8bitfw.py Checks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/mlx-community/Nanbeige4.1-3B-8bit.json --dir ./Nanbeige4.1-3B-8bit
$ python3 fw.py verify ./Nanbeige4.1-3B-8bit ./mlx-community__Nanbeige4.1-3B-8bit.jsonhf The Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT =https://freeweights.allternit.com hf download mlx-community/Nanbeige4.1-3B-8bit --revision 699b0e755a2841c4c2762f846d4eaf93542ac97c --local-dir ./Nanbeige4.1-3B-8bittransformers Load 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/Nanbeige4.1-3B-8bit", revision="699b0e755a2841c4c2762f846d4eaf93542ac97c")
model = AutoModelForCausalLM.from_pretrained("mlx-community/Nanbeige4.1-3B-8bit", revision="699b0e755a2841c4c2762f846d4eaf93542ac97c", device_map="auto")vLLM Serve an OpenAI-compatible API.
$ HF_ENDPOINT =https://freeweights.allternit.com vllm serve mlx-community/Nanbeige4.1-3B-8bit --revision 699b0e755a2841c4c2762f846d4eaf93542ac97cLicense Name Apache License 2.0 Id Apache-2.0 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 d9730bb8db566f224a41e93a666471133a84f5ffbcb54e0c3905ae05da06db2a 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: <huggingface.co/mlx-community> Newsgroups: alt.freeweights.textSubject: Nanbeige4.1-3B-8bit Date: 2026-11-15Commit: 699b0e7Signed: 2DC28CC7B26A283B ✓ Upstream mlx-community/Nanbeige4.1-3B-8bit Pinned commit 699b0e755a2841c4c2762f846d4eaf93542ac97c Manifest Nanbeige4.1-3B-8bit.json · .minisig Manifest sha256 f54489e9aecc31b53e000bc259e714ff1b2470d587d27ce93f76f9a167359a1c Signature minisign, Ed25519, key 2DC28CC7B26A283B Created 2026-11-15 05:04 UTC Generator freeweights-manifest/0.1.0 Takedowns None. How to verify
$ curl -O https://freeweights.allternit.com/manifests/mlx-community/Nanbeige4.1-3B-8bit.json
$ curl -O https://freeweights.allternit.com/manifests/mlx-community/Nanbeige4.1-3B-8bit.json.minisig
$ minisign -Vm Nanbeige4.1-3B-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 No model in the catalog names this one as its parent.
Get it
$ HF_ENDPOINT =https://freeweights.allternit.com hf download mlx-community/Nanbeige4.1-3B-8bitOpen in Freeweights Studio Pull, verify and run it in the Allternit app on Desktop, web or phone.
32.8B params 65.5 GB a2-rl-e2egit_large-50-32B safetensors permissive Facts Params 32.8B Size 65.5 GB Formats safetensors Task text generation License Apache-2.0 permissive flip From: laion <huggingface.co/laion> Subject: a2-rl-e2egit_large-50-32B 8.2B params 16.5 GB a3-rl-DCAgent_exp_rpt_e2egit-large-15-8B safetensors permissive Facts Params 8.2B Size 16.5 GB Formats safetensors Task not tagged License MIT permissive flip From: laion <huggingface.co/laion> Subject: a3-rl-DCAgent_exp_rpt_e2egit-large-15-8B 8.2B params 16.4 GB a3-rl-DCAgent_exp_rpt_e2egit-v2-10-8B safetensors permissive Facts Params 8.2B Size 16.4 GB Formats safetensors Task reinforcement learning License Apache-2.0 permissive flip From: laion <huggingface.co/laion> Subject: a3-rl-DCAgent_exp_rpt_e2egit-v2-10-8B 8.2B params 16.6 GB a3-rl-DCAgent_exp_rpt_pymethods2test-large-80-8B safetensors permissive Facts Params 8.2B Size 16.6 GB Formats safetensors Task not tagged License Apache-2.0 permissive flip From: laion <huggingface.co/laion> Subject: a3-rl-DCAgent_exp_rpt_pymethods2test-large-80-8B 8.2B params 16.6 GB a3-rl-DCAgent_exp_rpt_unitsyn-python-large-10-8B safetensors permissive Facts Params 8.2B Size 16.6 GB Formats safetensors Task not tagged License Apache-2.0 permissive flip From: laion <huggingface.co/laion> Subject: a3-rl-DCAgent_exp_rpt_unitsyn-python-large-10-8B 8.2B params 16.7 GB a3-rl-DCAgent_mix_h2_language_proportional-65-8B safetensors permissive Facts Params 8.2B Size 16.7 GB Formats safetensors Task not tagged License Apache-2.0 permissive flip From: laion <huggingface.co/laion> Subject: a3-rl-DCAgent_mix_h2_language_proportional-65-8B
mlx-community <huggingface.co/mlx-community>Newsgroups: alt.freeweights.textSubject: 
