nvidia/Qwen3.8-Flash-Next-NVFP4 — new model trending #28 on Hugging Face

# Model Overview

## Description:

The NVIDIA Qwen3.8-Flash-Next NVFP4 model is the quantized version of Alibaba's Qwen3.8-Flash-Next model, which is an auto-regressive language model that uses an optimized transformer architecture. Qwen3.8-Flash-Next is a causal language model with a vision encoder, hybrid attention (Gated DeltaNet and Qwen Sparse Attention), Mixture-of-Experts, gated residual streams, and n-gram embeddings. For more information, please check [here](https://huggingface.co/Qwen/Qwen3.8-Flash-Next). The NVIDIA Qwen3.8-Flash-Next NVFP4 model is quantized with [Model Optimizer](https://github.com/NVIDIA/Model-Optimizer).

This model is ready for commercial or 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 [(Qwen3.8-Flash-Next) Model Card](https://huggingface.co/Qwen/Qwen3.8-Flash-Next) from Qwen.

### License/Terms of Use:

Governing Terms: Use of this model is governed by the [NVIDIA Open Model License](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/).

**ADDITIONAL INFORMATION** : [Qwen Community License 1.0](https://huggingface.co/Qwen/Qwen3.8-Flash-Next/blob/main/LICENSE). **Qwen3.8-Flash-Next** .

### Deployment 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 08/31/2026 via https://huggingface.co/nvidia/Qwen3.8-Flash-Next-NVFP4 <br>

## Evaluation

The accuracy benchmark results are presented in the table below:

<table>

<tr>

<td><strong>Precision</strong>

</td>

<td><strong>GPQA Diamond</strong>

</td>

<td><strong>HLE</strong>

</td>

<td><strong>τ²-Bench Telecom</strong>

</td>

<td><strong>MMMU Pro</strong>

</td>

<td><strong>SciCode</strong>

</td>

<td><strong>AA-LCR</strong>

</td>

<td><strong>IFBench</strong>

</td>

<td><strong>Omniscience</strong>

</td>

<td><strong>Terminal-Bench 2.1</strong>

</td>

</tr>

<tr>

<td>FP8

</td>

<td><strong>92.0</strong>

</td>

<td><strong>34.7</strong>

</td>

<td><strong>90.8</strong>

</td>

<td><strong>77.1</strong>

</td>

<td><strong>16.3</strong>

</td>

<td><strong>71.9</strong>

</td>

<td><strong>80.5</strong>

</td>

<td><strong>28.1</strong>

</td>

<td><strong>83.3</strong>

</td>

</tr>

<tr>

<td>NVFP4

</td>

<td><strong>91.5</strong>

</td>

<td><strong>35.4</strong>

</td>

<td><strong>90.1</strong>

</td>

<td><strong>78.3</strong>

</td>

<td><strong>18.8</strong>

</td>

<td><strong>74.1</strong>

</td>

<td><strong>81.0</strong>

</td>

<td><strong>27.6</strong>

</td>

<td><strong>82.9</strong>

</td>

</tr>

</table>

> Baseline: [Qwen3.8-Flash-Next-FP8](https://huggingface.co/Qwen/Qwen3.8-Flash-Next-FP8). Benchmarked with temperature=1.0, top_p=0.95, max_new_tokens=131072, and reasoning_effort=xhigh.

## References

NVIDIA Model Optimizer: https://github.com/NVIDIA/Model-Optimizer

## Model Architecture:

**Architecture Type:** Transformers <br>

**Network Architecture:** Hybrid Attention (Gated DeltaNet and Qwen Sparse Attention) with Mixture-of-Experts (`Qwen4ExpForConditionalGeneration`) <br>

**Number of Model Parameters:** 125B in total and 6B activated, plus 51B n-gram embedding and 4B MTP <br>

**This model was developed based on [Qwen3.8-Flash-Next](https://huggingface.co/Qwen/Qwen3.8-Flash-Next)** <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 natively and extensible up to 1,000,000 tokens. <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>

**Post-Training Quantization Toolkit:** <br>

* [NVIDIA Model Optimizer](https://github.com/NVIDIA/Model-Optimizer) v0.46.0 with [Transformers v5.16.0](https://github.com/huggingface/transformers/releases/tag/v5.16.0) or later for upstream Qwen4-Exp support <br>

**Supported Hardware Microarchitecture Compatibility:** <br>

* NVIDIA Blackwell B200 <br>

* NVIDIA Blackwell B300 <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 and was produced with nvidia-modelopt **v0.46.0**. <br>

## Training and Evaluation Datasets:

## Calibration Dataset:

**Link:** [cnn_dailymail](https://huggingface.co/datasets/abisee/cnn_dailymail), [Nemotron-Post-Training-Dataset-v2](https://huggingface.co/datasets/nvidia/Nemotron-Post-Training-Dataset-v2) <br>

Text extracted automatically; images, tables and formatting may be missing. Original: https://huggingface.co/nvidia/Qwen3.8-Flash-Next-NVFP4