No official torrent yet — requesting it tells the flock what to host next. You cannot seed a catalog-only page.
Cataloged from nvidia/Qwen3-Coder-480B-A35B-Instruct-NVFP4
All models by nvidia →The NVIDIA Qwen3-Coder-480B-A35B-Instruct NVFP4 model is the quantized version of Alibaba's Qwen3-Coder-480B-A35B-Instruct model, which is an auto-regressive language model that uses an optimized transformer architecture. For more information, please check here. The NVIDIA Qwen3-Coder-480B-A35B-Instruct FP4 model is quantized with Model Optimizer.
This model is ready for commercial/non-commercial use.
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-Coder-480B-A35B-Instruct) Model Card.
Global
Developers looking to take off-the-shelf, pre-quantized models for deployment in AI Agent systems, chatbots, RAG systems, and other AI-powered applications.
Hugging Face 08/22/2025 via https://huggingface.co/nvidia/Qwen3-Coder-480B-A35B-Instruct-FP4
Architecture Type: Transformers
Network Architecture: Qwen3-Coder-480B-A35B-Instruct
**This model was developed based on Qwen/Qwen3-Coder-480B-A35B-Instruct
** Number of model parameters: 480B
Input Type(s): Text
Input Format(s): String
Input Parameters: 1D (One-Dimensional): Sequences
Other Properties Related to Input: Context length up to 256K
Output Type(s): Text
Output Format: String
Output Parameters: 1D (One-Dimensional): Sequences
Other Properties Related to Output: N/A
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.
Supported Runtime Engine(s):
Supported Hardware Microarchitecture Compatibility:
Preferred Operating System(s):
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.
The model is quantized with nvidia-modelopt v0.41.0
** Link: cnn_dailymail, Nemotron-Post-Training-Dataset-v2
** Data Collection Method by dataset: Automated.
** Labeling method: Automated.
** Data Collection Method by dataset by Dataset: Undisclosed
** Labeling Method by dataset: Undisclosed
** Properties: Undisclosed
** Data Collection Method by dataset by Dataset: Undisclosed
** Labeling Method by dataset: Undisclosed
** Properties: Undisclosed
Acceleration Engine: TensorRT-LLM
Test Hardware: B200
This model was obtained by quantizing the weights and activations of Qwen3-Coder-480B-A35B-Instruct to FP4 data type, ready for inference with TensorRT-LLM. Only the weights and activations of the linear operators within transformer blocks 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 3.5x.
To deploy the quantized checkpoint with TensorRT-LLM LLM API, follow the sample codes below:
The accuracy benchmark results are presented in the table below:
Baseline: Qwen/Qwen3-Coder-480B-A35B-Instruct. Benchmarked with temperature=0.0, top_p=1.0e-05, max num tokens 16384
NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse.
Please report model quality, risk, security vulnerabilities or NVIDIA AI Concerns here.
from tensorrt_llm import LLM, SamplingParams
def main():
prompts = [ "Hello, my name is", "The president of the United States is", "The capital of France is", "The future of AI is", ] sampling_params = SamplingParams(temperature=0.8, top_p=0.95)
llm = LLM(model="nvidia/Qwen3-Coder-480B-A35B-Instruct-FP4", tensor_parallel_size=4)
outputs = llm.generate(prompts, sampling_params)
# Print the outputs. for output in outputs: prompt = output.prompt generated_text = output.outputs[0].text print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
# The entry point of the program needs to be protected for spawning processes.if __name__ == '__main__': main()