AtomicChat / Model details

MiniCPM-V-4.7-35B-A3B-Uncensored-GGUF

AtomicChat/MiniCPM-V-4.7-35B-A3B-Uncensored-GGUF

uncensoredfine-tuned

Read the original model card ↗ · Choosing a model

Family: Unknown. No supported family evidence recorded.

About this model

AtomicChat's repository packages an uncensored derivative of OpenBMB's MiniCPM-V 4.7 35B-A3B in GGUF format under Apache-2.0. The creator says one refusal direction, derived from the residual stream at the last prompt token, was projected out of the original weights; the same edit is offered as three baked quantized builds and a 106 MB rank-1 LoRA adapter for existing MiniCPM-V 4.7 GGUFs. According to the model card, held-out refusal rates fell from 96.3% to 0% in English and from 70% to 0% in Russian, while MMLU changed by 0.25 points; these are the creator's measurements, and refusal detection only examines reply openings. The baked builds include AD-Q4_K-IQ4_XS, AD-Q5_K-Q4_K, and AD-Q6_K layouts, with the projector kept unmodified. The card says vision was not tested after the edit and that other languages were not measured.

Description sources

https://huggingface.co/AtomicChat/MiniCPM-V-4.7-35B-A3B-Uncensored-GGUF/raw/e414fa9ce8ca51e5b01e78d9d5f5d07a9b374443/README.md ↗

Source-based description, not an independent evaluation. Check the linked sources for the creator’s claims and limitations.

Size and formats

Base model: openbmb/MiniCPM-V-4.7-35B-A3B
Parameters: Unknown
Main repository formats: GGUF
License tag: apache-2.0
Languages: Unknown
Last updated: 2026-10-26 · 11:37:14.000 UTC

What changed

Model types are creator-reported, not independently tested. These labels describe source metadata, not a guarantee of behavior.

uncensoredmodel name: MiniCPM-V-4.7-35B-A3B-Uncensored-GGUF
fine-tunedsource tag: base_model:finetune:openbmb/MiniCPM-V-4.7-35B-A3B

Family evidence

These groupings are source-reported. Explicit base identifiers take priority over generic architecture tags.

No supported family evidence recorded.

Available versions

Versions appear here only when they explicitly name this exact model as their source. Check each repository for available files.

Main repository files ↗GGUF
No other explicitly linked versions recorded.

Link to this listing

Made or quantized this model? Paste one of these into your model card or README so people can find this listing.

Markdown

[Listed on Custom Clankers](https://customclankers.com/models/atomicchat-minicpm-v-4-7-35b-a3b-uncensored-gguf/)

HTML

<a href="https://customclankers.com/models/atomicchat-minicpm-v-4-7-35b-a3b-uncensored-gguf/">Listed on Custom Clankers</a>

Plain text links only: no image, no script, nothing loaded from this site.

Customize or build on this model

Listed from public GitHub data, not tested by us. Read the code and license before you run anything.

Tools to customize it further

Make a smaller download (quantize)

llama.cpp ↗ ★ 132,313
Runs models locally and includes the tools behind most GGUF files. A conversion script turns Hugging Face weights into GGUF, and llama-quantize shrinks a high-precision GGUF to lower-bit formats, which can cost some accuracy.

Learn the techniques

LLM Course ↗ ★ 84,684
A free course with roadmaps and Colab notebooks. Its notebooks include one-click tools for merging with MergeKit, quantizing to GGUF, GPTQ, EXL2, AWQ and HQQ, and automatic abliteration with custom datasets.

Compare all tools →

Source provenance

Downloads: 33,047
Likes: 42
Metadata checked: 2026-10-26 · 12:12:53 UTC
First seen in index: 2026-10-26 · 12:12:53 UTC
Source revision: e414fa9ce8ca51e5b01e78d9d5f5d07a9b374443

Source-card excerpt

Source model card excerpt: How to Run MiniCPM-V 4.7 35B-A3B Without Refusals Locally One refusal direction projected out of OpenBMB's original weights, shipped two ways: three baked GGUF

From AtomicChat/MiniCPM-V-4.7-35B-A3B-Uncensored-GGUF. Creator model-card excerpt, not an independent review.

Inspect before you run

Low refusal is not proof of correctness. Outputs can be inaccurate, harmful, or biased. Review the model card, license and files yourself. A license tag is not commercial-use advice. Parameter counts are source metadata; no hardware compatibility is implied.

This directory does not host or redistribute model weights. Refusal profiles are measured by us on a fixed prompt set; all other behavior claims are source-reported.

Collection rules & limitations ↗