Instructions to use hustvl/InfiniteVL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hustvl/InfiniteVL with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="hustvl/InfiniteVL", trust_remote_code=True)
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("hustvl/InfiniteVL", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use hustvl/InfiniteVL with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hustvl/InfiniteVL"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hustvl/InfiniteVL",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/hustvl/InfiniteVL

SGLang

How to use hustvl/InfiniteVL with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "hustvl/InfiniteVL" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hustvl/InfiniteVL",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "hustvl/InfiniteVL" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hustvl/InfiniteVL",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use hustvl/InfiniteVL with Docker Model Runner:
```
docker model run hf.co/hustvl/InfiniteVL
```

HongyuanTao commited on Dec 3, 2025

Commit

214835f

verified ·

1 Parent(s): 2714845

Update README.md

Browse files

Files changed (1) hide show

README.md +45 -3

README.md CHANGED Viewed

@@ -1,3 +1,45 @@
----
-license: apache-2.0
----

+---
+license: apache-2.0
+library_name: transformers
+tags:
+- vision-language-model
+- image-text-to-text
+- linear-attention
+- gated-deltanet
+- infinitevl
+- multimodal
+base_model: Qwen/Qwen2.5-VL-3B-Instruct
+pipeline_tag: image-text-to-text
+---
+<div align="center">
+# InfiniteVL: Synergizing Linear and Sparse Attention for Highly-Efficient, Unlimited-Input VLMs
+<a href="https://arxiv.org/abs/YOUR_ARXIV_ID"><img src="https://img.shields.io/badge/Paper-ArXiv-b31b1b.svg" alt="Paper"></a>
+<a href="https://github.com/YOUR_USERNAME/InfiniteVL"><img src="https://img.shields.io/badge/GitHub-Code-black" alt="Code"></a>
+<a href="LICENSE"><img src="https://img.shields.io/badge/License-Apache%202.0-blue.svg" alt="License"></a>
+</div>
+## 📖 Introduction
+**InfiniteVL** is a linear-complexity Vision-Language Model (VLM) developed by **Huazhong University of Science and Technology (HUST)** and **Horizon Robotics**.
+Traditional Transformer-based VLMs suffer from quadratic computational complexity ($O(N^2)$) and growing KV-cache memory usage. **InfiniteVL** solves this by synergizing **Sliding Window Attention (SWA)** with **Gated DeltaNet**, enabling **unlimited input tokens** and **real-time streaming**.
+### Key Features
+*   **🚀 Linear Complexity ($O(N)$):** Reduces per-token latency by **3.6×** compared to Qwen2.5-VL-3B.
+*   **📉 Constant Memory:** Maintains a fixed GPU memory usage (~9GB) regardless of sequence length.
+*   **⚡ Real-Time Streaming:** Sustains a stable **24 FPS** throughput for long video understanding on a single RTX 4090.
+*   **🧠 Hybrid Architecture:** 75% Gated DeltaNet (Global Context) + 25% SWA (Local Detail).
+![Performance Comparison](teaser.png)
+## 🛠️ Requirements
+To use InfiniteVL, you need to install the linear attention kernels.
+```bash
+pip install transformers torch
+pip install fla  # Flash Linear Attention