README.md

---
pipeline_tag: text-generation
base_model:
- zai-org/GLM-5
license: mit
library_name: Model Optimizer
tags:
- nvidia
- ModelOpt
- GLM-5
- quantized
- FP4
- fp4
---

# Model Overview

## Description:
The NVIDIA GLM-5 NVFP4 model is the quantized version of ZAI’s GLM-5 model, which is an auto-regressive language model that uses an optimized transformer architecture. For more information, please check [here](https://huggingface.co/zai-org/GLM-5). The NVIDIA GLM-5 NVFP4 model is quantized with [Model Optimizer](https://github.com/NVIDIA/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 [(GLM-5) Model Card](https://huggingface.co/zai-org/GLM-5) from ZAI.

## References
Nvidia Model Optimizer: https://github.com/NVIDIA/Model-Optimizer

### License/Terms of Use:
[MIT License](https://huggingface.co/datasets/choosealicense/licenses/blob/main/markdown/mit.md)

### 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>
Huggingface 03/16/2026 via https://huggingface.co/nvidia/GLM-5-NVFP4 <br>

## Model Architecture:
**Architecture Type:** Transformers  <br>
**Network Architecture:** GLM-5 <br>
**Number of Model Parameters:** 744B in total and 40B activated <br>

## Input:
**Input Type(s):** Text <br>
**Input Format(s):** String <br>
**Input Parameters:** One-Dimensional (1D) <br>
**Other Properties Related to Input:** Context length up to 200K <br>

## Output:
**Output Type(s):** Text <br>
**Output Format:** String <br>
**Output Parameters:** 1D (One-Dimensional): Sequences <br>
**Other Properties Related to Output:** N/A <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>
* SGLang <br>

**Supported Hardware Microarchitecture Compatibility:** <br>
* 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 **v0.42.0**  <br>

## Training, Testing, 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>
** Data Collection Method by dataset: Automated. <br>
** Labeling method: 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. <br>

## Training Dataset:
** Data Modality: Undisclosed <br>
** Data Collection Method by dataset: Undisclosed <br>
** Labeling Method by dataset: Undisclosed<br>
** Properties: Undisclosed

## Testing Dataset:
** Data Collection Method by dataset: Undisclosed <br>
** Labeling Method by dataset: Undisclosed <br>
** Properties: Undisclosed <br>

## Evaluation Dataset:
** Data Collection Method by dataset: Hybrid: Human, Automated <br>
** Labeling Method by dataset: Hybrid: Human, Automated <br>
** Properties: We evaluated the model on benchmarks including GPQA, which is a dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry. <br>


## Inference:
**Acceleration Engine:** SGLang <br>
**Test Hardware:** B300 <br>

## Post Training Quantization
This model was obtained by quantizing the weights and activations of GLM-5 to NVFP4 data type, ready for inference with SGLang. Only the weights and activations of the linear operators within transformer blocks in MoE are quantized.

## Usage

To serve this checkpoint with [SGLang](https://github.com/sgl-project/sglang), you can start the docker `lmsysorg/sglang:nightly-dev-cu13-20260305-33c92732` and run the sample command below:

```sh
python3 -m sglang.launch_server --model nvidia/GLM-5-NVFP4 --tensor-parallel-size 8 --quantization modelopt_fp4 --tool-call-parser glm47 --reasoning-parser glm45 --trust-remote-code --chunked-prefill-size 131072  --mem-fraction-static 0.80
```

If you would like to enable expert parallel when launch the SGLang endpoint, please build docker with provided [dockerfile](https://huggingface.co/nvidia/GLM-5-NVFP4/blob/main/dockerfile).

## Evaluation
The accuracy benchmark results are presented in the table below:
<table>
  <tr>
   <td><strong>Precision</strong>
   </td>
   <td><strong>MMLU Pro</strong>
   </td>
   <td><strong>GPQA Diamond</strong>
   </td>
   <td><strong>SciCode</strong>
   </td>
   <td><strong>IFBench</strong>
   </td>
   <td><strong>HLE</strong>
   </td>
  </tr>
  <tr>
   <td>FP8
   </td>
   <td>0.858
   </td>
   <td>0.862
   </td>
   <td>0.488
   </td>
   <td>0.717
   </td>
   <td>0.274
   </td>
  </tr>
  <tr>
   <td>NVFP4
   </td>
   <td>0.861
   </td>
   <td>0.855
   </td>
   <td>0.478
   </td>
   <td>0.712
   </td>
   <td>0.275
   </td>
  </tr>
  <tr>
</table>

> Baseline: [GLM-5-FP8](https://huggingface.co/zai-org/GLM-5-FP8).
> Benchmarked with temperature=1.0, top_p=0.95, max num tokens 131072

## Model Limitations:
The base model was trained on data that contains toxic language and societal biases originally crawled from the internet. Therefore, the model may amplify those biases and return toxic responses especially when prompted with toxic prompts. The model may generate answers that may be inaccurate, omit key information, or include irrelevant or redundant text producing socially unacceptable or undesirable text, even if the prompt itself does not include anything explicitly offensive.

## Ethical Considerations

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](https://app.intigriti.com/programs/nvidia/nvidiavdp/detail).