← back to catalog · registered 2026-08-22 13:56

vanch007/Huihui-MiniCPM-V-4.6-abliterated-mlx-bf16

vanch007 1.3B multimodal second-order
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/vanch007%2FHuihui-MiniCPM-V-4.6-abliterated-mlx-bf16"
Response includes
  • classification m1
  • files 11
  • hub_downloads_all_time 939
  • author_summary 27 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
Downloads · lifetime
939
69 last 30d - cooling
Likes
2
Model age
5mo ago
created 2026-05-13
Downloads over time
Now952→from63↑1,411%
193597001K63 on May 13952 on Oct 11MayJunJulAugSepOct
May 13 → Oct 11 · 61 snapshots · spans 151 days

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

Variants by this author 2 formats · 106 downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

Metadata

License
apache-2.0
Tags
mlx safetensors minicpmv4_6 minicpm-v multimodal image-text-to-text conversational base_model:huihui-ai/Huihui-MiniCPM-V-4.6-abliterated base_model:finetune:huihui-ai/Huihui-MiniCPM-V-4.6-abliterated license:apache-2.0 region:us

Related

Total size
2.42 GB
Files
11
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-05-13 17:39

Files by quantization

Auxiliary files 11 files 2.44 GB
model.safetensors 2.42 GB 6cc173be download
tokenizer.json 19.1 MB 83fd10d1 download
model.safetensors.index.json 63.8 KB c700a8d4 download
chat_template.jinja 7.13 KB f25b6ac3 download
config.json 2.71 KB 364317a6 download
.gitattributes 1.53 KB 52373fe2 download
tokenizer_config.json 1.17 KB 803b4997 download
README.md 986 B 036498cb download
processor_config.json 860 B a886452f download
preprocessor_config.json 315 B f8619291 download
generation_config.json 214 B b80dd7cb download

README current version from Hugging Face


license: apache-2.0
pipeline_tag: image-text-to-text
tags:

  • minicpm-v
  • multimodal
  • mlx
    library_name: mlx
    base_model: huihui-ai/Huihui-MiniCPM-V-4.6-abliterated

Huihui-MiniCPM-V-4.6-abliterated-mlx-bf16

Converted from huihui-ai/Huihui-MiniCPM-V-4.6-abliterated
with mlx-vlm from the current main branch.

Conversion

  • Source: huihui-ai/Huihui-MiniCPM-V-4.6-abliterated
  • Format: MLX
  • Quantization: bfloat16
  • Processor files: processor_config.json + preprocessor_config.json

Smoke test

  • Prompt: Describe this image in one sentence.
  • Image: 64x64 solid red PNG
  • Result: The image is a solid red color, with no discernible features or objects.

Usage

python -m mlx_vlm.generate --model vanch007/Huihui-MiniCPM-V-4.6-abliterated-mlx-bf16 --max-tokens 32 --temperature 0.0 --trust-remote-code --prompt "Describe this image in one sentence." --image <path_to_image>

README history 1 version

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-05-13Upload bf16 MLX conversion45eddc0986 B
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration