Ministral 3 14B Janus Abliterated V2 Base And GGUF is an open-source language model by SOMEHOTMEAL. Features: 14b LLM, VRAM: 28GB, Quantized, Instruction-Based, Merged, LLM Explorer Score: 0.32.
| LLM Name | Ministral 3 14B Janus Abliterated V2 Base And GGUF |
| Repository 🤗 | https://huggingface.co/SOMEHOTMEAL/Ministral-3-14B-Janus-Abliterated-V2-Base_And_GGUF |
| Base Model(s) | |
| Merged Model | Yes |
| Model Size | 14b |
| Required VRAM | 28 GB |
| Updated | 2026-09-16 |
| Maintainer | SOMEHOTMEAL |
| Model Type | mistral3 |
| Instruction-Based | Yes |
| Model Files | |
| GGUF Quantization | Yes |
| Quantization Type | fp16|gguf|q2|q8_0|q4_k |
| Model Architecture | Mistral3ForConditionalGeneration |
| Transformers Version | 5.5.3 |
| Tokenizer Class | TokenizersBackend |
| Padding Token | <pad> |
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Amaretto 14B | 0K / 28 GB | 714 | 3 |
| ... 3 14B Instruct 2512 MLX MXFP4 | 0K / 7.2 GB | 526 | 2 |
| Ministral 3 14B Abliterated | 0K / 28 GB | 46 | 5 |
| Ministral 3 14B Instruct 2512 | 0K / 27 GB | 60 | 1 |