From: mlx-community <huggingface.co/mlx-community> Newsgroups: alt.freeweights.minicpmSubject: MiniCPM5-2B-8bit Date: 2026-11-15Commit: 2d20e8eSigned: 2DC28CC7B26A283B ✓ Text generation model with 2.5B parameters. Safetensors weights, 2.68 GB in all. Published by mlx-community under Apache-2.0.
2.5B params 2.68 GB MiniCPM5-2B-8bit safetensors permissive Facts Params 2.5B Size 2.68 GB Formats safetensors Task text generation License Apache-2.0 From sourcepermissive
fw
$ fw get mlx-community/MiniCPM5-2B-8bit
# manifest sha256 ec99afff4513… · 2.68 GB · 9 files
$ fw verify ./MiniCPM5-2B-8bit mlx-community/MiniCPM5-2B-8bit
From sourceHosted (signed) mlx-community/MiniCPM5-2B-8bit Last check: 2026-11-15 05:02 UTC
Get help: custom fine-tune or private deploy. We review your setup before we quote.
Quantized of MiniCPM5-2B .
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 2.68 GB
Weights 2.67 GB (safetensors)
Parameters 2.5B
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.minicpm
Lineage
Original release from the publisher.
Base model: openbmb/MiniCPM5-2B
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 From sourcehuggingface.co 2026-11-15 05:02 UTC passed
Model card From the publisher at commit
2d20e8e . Shown as plain text with links; images open at the publisher.
This model mlx-community/MiniCPM5-2B-8bit was converted to MLX format from openbmb/MiniCPM5-2B using mlx-lm version 0.31.3 (8-bit, group size 64).
Use with mlx#
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/MiniCPM5-2B-8bit")
prompt = "Hello, who are you?"
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) Copy Files 9 files · 2.68 GB Weights 2.67 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 9 files with their sha256 File Size sha256 .gitattributes 1.52 KB 11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361 README.md 1.20 KB 55f0015faf3ff4e9511670aefab659780b855b288082e49bbddf92151ff9b603 chat_template.jinja 9.06 KB cc945752db555d60949b16989df4ccfeb52a313d6b4b5c5229dd786e2e9fcf1c config.json 886 B aa0fa827a83a29f2bc8b1b9c681914b0689554762f11f3d7240725133f8a104f generation_config.json 213 B 9ac4f32e5f32358697a9f438a3ea89ef80e6ba786c72c49e932f9f21c122fdb1 model.safetensors 2.67 GB 82db8e423c5df99a5989f1b73dad1f44941a7800b308915f6ae7142966b32e40 model.safetensors.index.json 68.7 KB 50747e37090508f851cdea2f4d5a71c59c2a26a09c8e95c505371ac669e2ac36 tokenizer.json 9.89 MB 3e065a558a034185fe299917b398685c1facd0169a9eea1e629eb30c171fed81 tokenizer_config.json 435 B b89503c3e5070c6b6d33daf2e20cb4a5c88537c1670d9b7e0cfb4506a61448a9
Versions Each time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commit Date Contents 2d20e8e672 pinned now 2026-11-15 05:02 UTC 9 files · 2.68 GB
Use it fw
$ fw get mlx-community/MiniCPM5-2B-8bit
# manifest sha256 ec99afff4513… · 2.68 GB · 9 files
$ fw verify ./MiniCPM5-2B-8bit mlx-community/MiniCPM5-2B-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/MiniCPM5-2B-8bit.json --dir ./MiniCPM5-2B-8bit
$ python3 fw.py verify ./MiniCPM5-2B-8bit ./mlx-community__MiniCPM5-2B-8bit.json Copyhf The Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT =https://freeweights.allternit.com hf download mlx-community/MiniCPM5-2B-8bit --revision 2d20e8e672ce892d50f7265bfd3fc9b59b718f2a --local-dir ./MiniCPM5-2B-8bit Copytransformers 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/MiniCPM5-2B-8bit", revision="2d20e8e672ce892d50f7265bfd3fc9b59b718f2a")
model = AutoModelForCausalLM.from_pretrained("mlx-community/MiniCPM5-2B-8bit", revision="2d20e8e672ce892d50f7265bfd3fc9b59b718f2a", device_map="auto") CopyvLLM Serve an OpenAI-compatible API.
$ HF_ENDPOINT =https://freeweights.allternit.com vllm serve mlx-community/MiniCPM5-2B-8bit --revision 2d20e8e672ce892d50f7265bfd3fc9b59b718f2a CopyLicense 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 55f0015faf3ff4e9511670aefab659780b855b288082e49bbddf92151ff9b603 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.minicpmSubject: MiniCPM5-2B-8bit Date: 2026-11-15Commit: 2d20e8eSigned: 2DC28CC7B26A283B ✓ Upstream mlx-community/MiniCPM5-2B-8bit Pinned commit 2d20e8e672ce892d50f7265bfd3fc9b59b718f2a Manifest MiniCPM5-2B-8bit.json · .minisig Manifest sha256 ec99afff45134ef57089368567294574596374321ac8c0d9f87fa82b0f01431d Signature minisign, Ed25519, key 2DC28CC7B26A283B Created 2026-11-15 05:02 UTC Generator freeweights-manifest/0.1.0 Takedowns None.
How to verify
$ curl -O https://freeweights.allternit.com/manifests/mlx-community/MiniCPM5-2B-8bit.json
$ curl -O https://freeweights.allternit.com/manifests/mlx-community/MiniCPM5-2B-8bit.json.minisig
$ minisign -Vm MiniCPM5-2B-8bit.json -P RWQ7KGqyx4zCLXtUFOqvfv8L3g+8fpKJ1FsuxPyOYMF+4KufbUxKq1Oq
$ shasum -a 256 <file> # compare with the Files tab Copy
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
Download
$ HF_ENDPOINT =https://freeweights.allternit.com hf download mlx-community/MiniCPM5-2B-8bit Copy
Open in Freeweights Studio Pull, verify and run it in the Allternit app on Desktop, web or phone.
minicpm 20.6 GB AgentCPM-Explore-GGUF gguf permissive Facts Params not stated Size 20.6 GB Formats gguf Task not tagged License Apache-2.0 From sourcepermissive flip From: openbmb <huggingface.co/openbmb> Subject: AgentCPM-Explore-GGUF 8.2B params 16.4 GB AgentCPM-Report safetensors permissive Facts Params 8.2B Size 16.4 GB Formats safetensors Task text generation License Apache-2.0 From sourcepermissive flip From: openbmb <huggingface.co/openbmb> Subject: AgentCPM-Report minicpm 4.97 GB AgentCPM-Report-GGUF gguf permissive Facts Params not stated Size 4.97 GB Formats gguf Task not tagged License Apache-2.0 From sourcepermissive flip From: openbmb <huggingface.co/openbmb> Subject: AgentCPM-Report-GGUF minicpm 1.11 GB BitCPM-CANN-0.5B-gguf gguf permissive Facts Params not stated Size 1.11 GB Formats gguf Task text generation License Apache-2.0 From sourcepermissive flip From: openbmb <huggingface.co/openbmb> Subject: BitCPM-CANN-0.5B-gguf minicpm 3.80 GB BitCPM-CANN-1B-gguf gguf permissive Facts Params not stated Size 3.80 GB Formats gguf Task text generation License Apache-2.0 From sourcepermissive flip From: openbmb <huggingface.co/openbmb> Subject: BitCPM-CANN-1B-gguf minicpm 8.31 GB BitCPM-CANN-3B-gguf gguf permissive Facts Params not stated Size 8.31 GB Formats gguf Task text generation License Apache-2.0 From sourcepermissive flip From: openbmb <huggingface.co/openbmb> Subject: BitCPM-CANN-3B-gguf