Qwen3.6 35B A3B NVFP4
Overview
The Qwen3.6 35B A3B NVFP4 is a specialized iteration of the Qwen series, optimized specifically for NVIDIA hardware using the NVFP4 quantization format. For developers, the primary draw here is the efficiency gain; by leveraging 4-bit floating point precision, this model significantly reduces VRAM overhead without the drastic perplexity loss typically seen in integer quantization. It is designed for high-throughput text generation and complex reasoning tasks where latency is critical. Integration is streamlined for NVIDIA TensorRT-LLM environments, making it an ideal candidate for production-grade RAG pipelines or agentic workflows where you need the intelligence of a mid-sized model but the speed of a much smaller one.
Highlights
- Optimized for NVIDIA hardware via NVFP4 precision
- Reduced VRAM footprint with minimal accuracy loss
- High-throughput performance for real-time text generation
- Apache-2.0 license for flexible commercial deployment
- Seamless integration with TensorRT-LLM frameworks
Usage
Install
# Install Hugging Face transformers
pip install transformers torchSDK Usage
# Load model with transformers
from transformers import AutoModel, AutoTokenizer
model = AutoModel.from_pretrained("nvidia/Qwen3.6-35B-A3B-NVFP4")
tokenizer = AutoTokenizer.from_pretrained("nvidia/Qwen3.6-35B-A3B-NVFP4")Hugging Face Download
We recommend downloading the model via the Hugging Face CLI or Hub SDK.
Guidance：Before downloading, install huggingface_hub with:
Guidance
pip install -U huggingface_hubCLI Download
Download the full repository
Download the full repository
huggingface-cli download nvidia/Qwen3.6-35B-A3B-NVFP4Download a single file to a local folder (e.g. config.json into ./dir)
Download a single file to a local folder (e.g. config.json into ./dir)
huggingface-cli download nvidia/Qwen3.6-35B-A3B-NVFP4 config.json --local-dir ./dirSee the official docs for more CLI options
SDK Download
SDK Download
# 模型下载
from huggingface_hub import snapshot_download
model_dir = snapshot_download('nvidia/Qwen3.6-35B-A3B-NVFP4')Git Download
Make sure git-lfs is installed first
Git Download
git lfs install
git clone https://huggingface.co/nvidia/Qwen3.6-35B-A3B-NVFP4To skip LFS large-file downloads, use:
Skip LFS
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/nvidia/Qwen3.6-35B-A3B-NVFP4Model files are hosted on the Hugging Face Hub — download directly via HF CLI / SDK / Git, not through this site.
PyTorch / Transformers Usage
Install Transformers
Install Transformers
pip install -U transformers torchLoad the model and run inference
Load the model and run inference
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained('nvidia/Qwen3.6-35B-A3B-NVFP4')
tokenizer = AutoTokenizer.from_pretrained('nvidia/Qwen3.6-35B-A3B-NVFP4')Full Documentation
来源: HuggingFace
---
pipeline_tag: text-generation
base_model:
- Qwen/Qwen3.6-35B-A3B
license: apache-2.0
library_name: Model Optimizer
tags:
- nvidia
- ModelOpt
- Qwen3.6
- quantized
- FP4
- fp4
---
Model Overview
Description:
The NVIDIA Qwen3.6-35B-A3B-NVFP4 model is the quantized version of Alibaba's Qwen3.6-35B-A3B model, which is an auto-regressive language model that uses an optimized transformer architecture. For more information, please check here. The NVIDIA Qwen3.6-35B-A3B-NVFP4 model is quantized with Model Optimizer.This model is ready for commercial/non-commercial use. <br>
Third-Party Community Consideration
This model is not owned or developed by NVIDIA. This model has been developed and built to a third-party’s requirements for this application and use case; see link to Non-NVIDIA (Qwen3.6-35B-A3B) Model Card from Alibaba.References
NVIDIA Model Optimizer: https://github.com/NVIDIA/Model-OptimizerLicense/Terms of Use:
Apache license 2.0Deployment Geography:
Global <br>Use Case: <br>
Developers looking to take off-the-shelf, pre-quantized models for deployment in AI Agent systems, chatbots, RAG systems, and other AI-powered applications. <br>Release Date: <br>
Hugging Face on 05/28/2026 via https://huggingface.co/nvidia/Qwen3.6-35B-A3B-NVFP4 <br>Model Architecture:
Architecture Type: Transformers <br> Network Architecture: Mixture-of-Experts (MoE) with Hybrid Attention <br> Number of Model Parameters: 35B in total and 3B activated <br>Input:
Input Type(s): Text, Image, Video <br> Input Format(s): String, Red, Green, Blue (RGB), Video (MP4/WebM) <br> Input Parameters: One-Dimensional (1D), Two-Dimensional (2D), Three-Dimensional (3D) <br> Other Properties Related to Input: Context length up to 262K <br>Output:
Output Type(s): Text <br> Output Format: String <br> Output Parameters: One-Dimensional(1D): Sequences <br> Other Properties Related to Output: None <br>Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA’s hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions. <br>
Software Integration:
Supported Runtime Engine(s): <br>- vLLM <br>
Supported Hardware Microarchitecture Compatibility: <br>
- NVIDIA Hopper, NVIDIA Blackwell <br>
Preferred Operating System(s): <br>
- Linux <br>
The integration of foundation and fine-tuned models into AI systems requires additional testing using use-case-specific data to ensure safe and effective deployment. Following the V-model methodology, iterative testing and validation at both unit and system levels are essential to mitigate risks, meet technical and functional requirements, and ensure compliance with safety and ethical standards before deployment.
Model Version(s):
The model version is NVFP4 1.0 version and is Quantized with nvidia-modelopt v0.44.0 <br>Training and Evaluation Datasets:
Calibration Dataset:
Link: cnn_dailymail, Nemotron-Post-Training-Dataset-v2 <br> Data Collection Method by dataset: Automated. <br> Labeling Method by dataset: Automated. <br> Properties: The cnn_dailymail dataset is an English-language dataset containing just over 300k unique news articles as written by journalists at CNN and the Daily Mail. The Nemotron-Post-Training-Dataset-v2 is a post-training dataset curated by NVIDIA containing multi-turn conversations across diverse topics. <br>Training Dataset:
Data Modality: Undisclosed <br> Data Collection Method by dataset: Undisclosed <br> Labeling Method by dataset: Undisclosed <br> Data Size: Undisclosed <br> Properties: UndisclosedEvaluation Dataset:
Datasets: MMLU Pro, GPQA Diamond, τ²-Bench Telecom, MMMU Pro, SciCode, AIME 2025, AA-LCR, IFBench <br> Data Collection Method by dataset: Hybrid: Automated, Human <br> Labeling Method by dataset: Hybrid: Human, Automated <br> Properties: We evaluated the model on text-based reasoning and coding benchmarks: MMLU Pro is a multi-task language understanding benchmark with challenging multiple-choice questions across diverse academic domains; GPQA Diamond contains 448 graduate-level multiple-choice questions written by domain experts in biology, physics, and chemistry; τ²-Bench Telecom evaluates agentic tool-use and policy-adherence capabilities in dual-control telecom customer-service scenarios where the model interacts with a simulated user and external tools to resolve account issues; MMMU Pro is the more challenging version of the Massive Multi-discipline Multimodal Understanding benchmark, measuring college-level multimodal reasoning across diverse disciplines with expanded answer choices and a vision-only input setting; SciCode evaluates scientific coding capabilities; AIME 2025 contains problems from the American Invitational Mathematics Examination; AA-LCR (Artificial Analysis Long Context Recall) evaluates a model's ability to accurately retrieve and recall information from long input contexts; IFBench is a benchmark for evaluating instruction-following capabilities across diverse and structured task constraints. <br>Inference:
Acceleration Engine: vLLM <br> Test Hardware: NVIDIA GB300 <br>Post Training Quantization
This model was obtained by quantizing the weights of Qwen3.6-35B-A3B to NVFP4 data type, ready for inference with vLLM. Only the weights and activations of the linear operators within transformer blocks in MoE are quantized. This optimization reduces the number of bits per parameter from 16 to 4, reducing the disk size and GPU memory requirements by approximately 3.06x.Usage
To serve this checkpoint with vLLM, you can start the docker vllm/vllm-openai:nightly and run the sample command below:
sh
vllm serve nvidia/Qwen3.6-35B-A3B-NVFP4 --port 8000 --quantization modelopt --max-model-len 262144 --reasoning-parser qwen3For NVIDIA DGX Spark, we recommend using this vllm serve command:
sh
vllm serve nvidia/Qwen3.6-35B-A3B-NVFP4 \
--host 0.0.0.0 \
--port 8000 \
--tensor-parallel-size 1 \
--trust-remote-code \
--kv-cache-dtype fp8 \
--attention-backend flashinfer \
--moe-backend marlin \
--gpu-memory-utilization 0.4 \
--max-model-len 262144 \
--max-num-seqs 4 \
--max-num-batched-tokens 8192 \
--enable-chunked-prefill \
--async-scheduling \
--enable-prefix-caching \
--speculative-config '{"method":"mtp","num_speculative_tokens":3,"moe_backend":"triton"}' \
--load-format fastsafetensors \
--reasoning-parser qwen3 \
--tool-call-parser qwen3_xml \
--enable-auto-tool-choice