Get the app
← back to catalog · registered 2026-09-13 16:56

SOMEHOTMEAL/Ministral-3-14B-Janus-Abliterated-V2-Base_And_GGUF

SOMEHOTMEAL 14B GGUF multimodal
Abliteration classifier · v1.0.0
M5
Primary method

Mergekit merge

Applied on top of direct removal inherited from the base model.
Confidence
MEDIUM
Inherited from base model
M1 Direct removal →
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.
  • merge tag / mergekit / dare-ties in tags or name
  • no unusual architecture pattern (regular merge)
  • abliterated marker present
How we label → About M5 →
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · 30-day
0
Likes
2
Model age
today
created 2026-09-13

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.

mistralai/Ministral-3-14B-Instruct-2512 this lineage
SOMEHOTMEAL/Ministral-3-14B-Janus-Abliterated-V2-Base_And_GGUF

Metadata

Quantizations
Q4_K
Tags
transformers safetensors gguf mistral3 image-text-to-text mergekit merge conversational arxiv:2311.03099 base_model:jenerallee78/Ministral-3-14B-abliterated base_model:merge:jenerallee78/Ministral-3-14B-abliterated base_model:mistralai/Ministral-3-14B-Instruct-2512

Related

Other by SOMEHOTMEAL
Qwen3-14B-Janus-Abliterated-GGUF 0 Qwen3.5-9B-Janus-Abliterated-GGUF 0 Qwen3-8B-Janus-Abliterated-GGUF 0 Gemma-4-12B-Janus-Abliterated-V2-Base_And_GGUF 0 Ministral-3-14B-Janus-Abliterated-GGUF 0
Same base model
jenerallee78/Ministral-3-14B-abliterated 20 HifiHu/Ministral-3-14B-abliterated 6
Total size
87.4 GB
Files
20
Quantizations
2
Registered
2026-09-13 16:56
Last updated on HF
2026-09-13 16:48
Source huggingface.co/SOMEHOTMEAL/Ministral-3-14B-Janus-Abliterated-V2-Base_And_GGUF ↗ Base model mistralai/Ministral-3-14B-Instruct-2512 ↗

Files by quantization

Q4_K 1 file 7.67 GB
Ministral-3-14B-Janus-Abliterated-V2-Q4_K_M-GGUF.gguf 7.67 GB 0a503cbc download
Auxiliary files 19 files 79.7 GB
Ministral-3-14B-Janus-Abliterated-V2-f16-GGUF.gguf 25.2 GB 7d8a169d download
Ministral-3-14B-Janus-Abliterated-V2-Q8_0-GGUF.gguf 13.4 GB 8dfa7f58 download
Ministral-3-14B-Janus-Abliterated-V2-Q6_K-GGUF.gguf 10.3 GB 5065ad57 download
Ministral-3-14B-Janus-Abliterated-V2-Q2_K-GGUF.gguf 4.89 GB b5acfefc download
model-00002-of-00006.safetensors 4.53 GB 705c57b6 download
model-00003-of-00006.safetensors 4.53 GB 78f03e4e download
model-00004-of-00006.safetensors 4.53 GB 881eeaa3 download
model-00005-of-00006.safetensors 4.53 GB c81be3b3 download
model-00001-of-00006.safetensors 4.51 GB 5c5eecb2 download
model-00006-of-00006.safetensors 3.34 GB 24092c03 download
tokenizer.json 16.3 MB 99cf2742 download
model.safetensors.index.json 57.2 KB bac1d3cb download
tokenizer_config.json 21.7 KB 90a7cc1c download
chat_template.jinja 11.8 KB dc2c84ef download
config.json 2.04 KB 00e9a72b download
.gitattributes 1.96 KB b749b008 download
README.md 1.08 KB 08db1b37 download
processor_config.json 976 B a37d728b download
mergekit_config.yml 243 B 784bbbd5 download

README current version from Hugging Face

base_model:

  • mistralai/Ministral-3-14B-Instruct-2512
  • jenerallee78/Ministral-3-14B-abliterated
    library_name: transformers
    tags:
  • mergekit
  • merge

Ministral-3-14B-Janus-Abliterated-V2

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the Linear DARE merge method using mistralai/Ministral-3-14B-Instruct-2512 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

merge_method: dare_linear
base_model: mistralai/Ministral-3-14B-Instruct-2512

models:
  - model: jenerallee78/Ministral-3-14B-abliterated
    parameters:
      weight: 0.60
      density: 0.70

dtype: bfloat16
tokenizer_source: base
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 app" button that hands off directly to a local runtime of your choice - Infrahuman, LM Studio, or Ollama. No API keys, no subscription, no prompt leakage.

See supported apps Browse the catalog