nvidia/EGM-4B-SFT

EGM-Qwen3-VL-4B-SFT

[Project Page]   [Code]  

Model Summary

EGM-Qwen3-VL-4B-SFT is the supervised fine-tuning (SFT) checkpoint from the first stage of the EGM (Efficient Visual Grounding Language Models) training pipeline. It is built on top of Qwen3-VL-4B-Thinking.

This is an intermediate checkpoint intended for further reinforcement learning training. For the final model with best performance, see nvidia/EGM-4B.

Training Details

SFT Stage

In the SFT stage, a proprietary VLM generates detailed chain-of-thought reasoning steps for visual grounding training data. The base Qwen3-VL-4B-Thinking model is then fine-tuned on this reasoning-augmented data to learn structured visual grounding with explicit reasoning.

This SFT checkpoint serves as the initialization for the subsequent RL stage (GRPO), which yields the final EGM-4B model.

How to Use for RL Training

pip install -U huggingface_hub
huggingface-cli download nvidia/EGM-4B-SFT --local-dir ./models/EGM-4B-SFT

Then follow the installation and RL training instructions in the EGM repository.

Model Architecture

Component	Details
Architecture	Qwen3VLForConditionalGeneration
Precision	bfloat16
Text Hidden Size	2560
Text Layers	36
Attention Heads	32 (8 KV heads)
Text Intermediate Size	9728
Vision Hidden Size	1024
Vision Layers	24
Patch Size	16 x 16
Max Position Embeddings	262,144
Vocabulary Size	151,936

Related Models

Model	Description
nvidia/EGM-4B	Final RL-trained model (best performance)
nvidia/EGM-8B-SFT	SFT checkpoint for the 8B variant
nvidia/EGM-8B	Final RL-trained 8B model

Citation

@article{zhan2026EGM,
    author = {Zhan, Guanqi and Li, Changye and Liu, Zhijian and Lu, Yao and Wu, Yi and Han, Song and Zhu, Ligeng},
    title = {EGM: Efficient Visual Grounding Language Models},
    booktitle = {arXiv},
    year = {2026}
}

Acknowledgment

This repository benefits from Qwen3-VL, InternVL, verl and verl-internvl.