Josiefied-Qwen3.5-0.8B-gabliterated-v1-MNN

Pre-converted Josiefied-Qwen3.5-0.8B-gabliterated-v1 in MNN format for on-device inference with TokForge.

Original model by Goekdeniz-Guelmez — converted to MNN Q4 for mobile deployment.

Model Details

Architecture	Qwen3.5 (hybrid attention: full + LinearAttention, 24 layers)
Parameters	0.8B (4-bit quantized)
Format	MNN (Alibaba Mobile Neural Network)
Quantization	W4A16 (4-bit weights, block size 128)
Vocab	248,320 tokens
Source	Goekdeniz-Guelmez/Josiefied-Qwen3.5-0.8B-gabliterated-v1

Description

Josiefied gabliterated 0.8B by Goekdeniz Guelmez — an ultra-compact uncensored Qwen3.5 model. Perfect for low-RAM devices (runs on 4GB+ phones) or as a fast draft model for speculative decoding. The "gabliterated" technique combines gradient-based and activation-based abliteration for robust uncensoring.

Files

File	Description
llm.mnn	Model computation graph
llm.mnn.weight	Quantized weight data (Q4, block=128)
llm_config.json	Model config with Jinja chat template
tokenizer.txt	Tokenizer vocabulary
config.json	MNN runtime config

Usage with TokForge

This model is optimized for TokForge — a free Android app for private, on-device LLM inference.

1. Download TokForge from the Play Store
2. Open the app → Models → Download this model
3. Start chatting — runs 100% locally, no internet required

Recommended Settings

Setting	Value
Backend	OpenCL (Qualcomm) / Vulkan (MediaTek) / CPU (fallback)
Precision	Low
Threads	4
Thinking	Off (or On for thinking-capable models)

Performance

Actual speed varies by device, thermal state, and generation length. Typical ranges for this model size:

Device	SoC	Backend	tok/s
SM8850 (RedMagic)	Snapdragon 8 Elite 2	CPU	~35 tok/s
SM8650 (Lenovo)	Snapdragon 8 Gen 3	CPU	~25 tok/s
SM8635 (Xiaomi)	Snapdragon 7+ Gen 3	CPU	~18 tok/s

Attribution

This is an MNN conversion of Josiefied-Qwen3.5-0.8B-gabliterated-v1 by Goekdeniz-Guelmez. All credit for the model architecture, training, and fine-tuning goes to the original author(s). This conversion only changes the runtime format for mobile deployment.

Limitations

• Intended for TokForge / MNN on-device inference on Android
• This is a runtime bundle, not a standard Transformers training checkpoint
• Quantization (Q4) may slightly reduce quality compared to the full-precision original
• Abliterated/uncensored models have had safety filters removed — use responsibly

Community

• Website: tokforge.ai
• Discord: Join our Discord
• GitHub: TokForge on GitHub

Export Details

Converted using MNN's llmexport pipeline:

python llmexport.py --path Goekdeniz-Guelmez/Josiefied-Qwen3.5-0.8B-gabliterated-v1 --export mnn --quant_bit 4 --quant_block 128