File	Quantization	Description
`Qwen3.5-35B-A3B-Opus-Reasoning-Distilled-Uncensored-bf16.gguf`	BF16	Full precision, suitable for further quantization
`Qwen3.5-35B-A3B-Opus-Reasoning-Distilled-Uncensored-Q4_K_M.gguf`	Q4_K_M	4-bit quantized, good quality/size balance

Usage with llama.cpp

bash

# Full precision
./llama-cli -m Qwen3.5-35B-A3B-Opus-Reasoning-Distilled-Uncensored-bf16.gguf -p "Your prompt here" -n 512

# Quantized (recommended for most users)
./llama-cli -m Qwen3.5-35B-A3B-Opus-Reasoning-Distilled-Uncensored-Q4_K_M.gguf -p "Your prompt here" -n 512

Usage with Ollama

Create a Modelfile:

FROM ./Qwen3.5-35B-A3B-Opus-Reasoning-Distilled-Uncensored-Q4_K_M.gguf

Then run:

bash

ollama create my-model -f Modelfile
ollama run my-model

Quantization Details

Tool: llama.cpp (convert_hf_to_gguf.py + llama-quantize)
Pipeline: HF safetensors -> bf16 GGUF -> Q4_K_M GGUF

ponytang3/Qwen3.5-35B-A3B-Opus-Reasoning-Distilled-GGUF

Qwen3.5-35B-A3B-Opus-Reasoning-Distilled-Uncensored-GGUF

Model Description

Available Files

Usage with llama.cpp

Usage with Ollama

Quantization Details

ponytang3/
Qwen3.5-35B-A3B-Opus-Reasoning-Distilled-GGUF