Qwen3.6-35B-A3B-NVFP4-GGUF

---

license: apache-2.0 language:

• en library_name: gguf tags:
• gguf
• qwen3.6
• qwen
• nvfp4
• blackwell
• fp4
• mixture-of-experts
• moe
• multimodal
• vision base_model: Qwen/Qwen3.6-35B-A3B pipeline_tag: image-text-to-text inference: false quantized_by: FreedomAISVR

---

Qwen3.6-35B-A3B-NVFP4-GGUF

NVFP4 GGUF quantization of Qwen/Qwen3.6-35B-A3B.

Multimodal model: vision encoder (903 MB) + text MoE LLM (18.36 GB).

About NVFP4

NVFP4 is a Blackwell-native FP4 format (EFM4: 1 sign, 1 mantissa, 2 exponent bits). Applied uniformly to all tensors.

• Total params: 35.95B (3B active, 256 experts, 8/token)
• Quantization: NVFP4 (~4.55 BPW)
• File size: 18.36 GiB (text) + 903 MB (vision)
• Vision encoder: 27-layer ViT, hidden 1152, 3->1152x2 patch embed (temporal)
• Context: 262,144 tokens natively

NVFP4 requires a Blackwell (RTX 50-series or B-series) GPU for hardware acceleration.

Files

Usage

llama.cpp CLI (text only)

llama-cli -hf FreedomAISVR/Qwen3.6-35B-A3B-NVFP4-GGUF -cnv -p "You are a helpful assistant"

llama-server (multimodal)

llama-server -hf FreedomAISVR/Qwen3.6-35B-A3B-NVFP4-GGUF --mmproj mmproj-qwen36-35b-src-BF16.gguf --ctx-size 0 --jinja

llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
    repo_id="FreedomAISVR/Qwen3.6-35B-A3B-NVFP4-GGUF",
    filename="qwen3.6-35b-a3b-nvfp4.gguf",
)

Quantization Pipeline

# 1. Convert HF model to intermediate GGUF
python convert_hf_to_gguf.py ./models/qwen3.6-35b/ --outfile qwen3.6-35b-a3b-f16.gguf --outtype bf16

# 2. Export vision encoder
python convert_hf_to_gguf.py ./models/qwen3.6-35b/ --mmproj --outtype bf16

# 3. Quantize to NVFP4
llama-quantize --allow-requantize qwen3.6-35b-a3b-f16.gguf qwen3.6-35b-a3b-nvfp4.gguf NVFP4

Hardware

License

Apache-2.0 (same as Qwen/Qwen3.6-35B-A3B)