Model Intelligence Sheet
Auguments/Josiefied-Qwen3.5-2B-gabliterated-v1-MTP-GGUF-BF16 overview
Josiefied Qwen3.5 2B gabliterated v1 Highest quality GGUF BF16. In this repo: BF16 conversion, and Q6 K converted from BF16 MTP block included For mmproj, use …
Runs locally from ~1.50 GB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
Model Details
Model README
---
base_model: Goekdeniz-Guelmez/Josiefied-Qwen3.5-2B-gabliterated-v1
base_model_relation: quantized
library_name: gguf
tags:
- gguf
- quantized
---
Josiefied-Qwen3.5-2B-gabliterated-v1
Highest quality GGUF - BF16.
In this repo:
- BF16 conversion, and Q6_K converted from BF16
- MTP block included
For mmproj, use unsloth's or bartowski's.
Converted using base llama.cpp b9294. Never versions can't convert this particular repo to GGUF due to regressed Transformers.
Run Auguments/Josiefied-Qwen3.5-2B-gabliterated-v1-MTP-GGUF-BF16 with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models