Instructions to use ACERobotics/ACE-Brain-0-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ACERobotics/ACE-Brain-0-8B with Transformers:

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

pipe = pipeline("image-text-to-text", model="ACERobotics/ACE-Brain-0-8B")
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 AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("ACERobotics/ACE-Brain-0-8B")
model = AutoModelForMultimodalLM.from_pretrained("ACERobotics/ACE-Brain-0-8B")
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?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ACERobotics/ACE-Brain-0-8B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ACERobotics/ACE-Brain-0-8B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ACERobotics/ACE-Brain-0-8B",
		"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/ACERobotics/ACE-Brain-0-8B

SGLang

How to use ACERobotics/ACE-Brain-0-8B 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 "ACERobotics/ACE-Brain-0-8B" \
    --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": "ACERobotics/ACE-Brain-0-8B",
		"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 "ACERobotics/ACE-Brain-0-8B" \
        --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": "ACERobotics/ACE-Brain-0-8B",
		"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 ACERobotics/ACE-Brain-0-8B with Docker Model Runner:
```
docker model run hf.co/ACERobotics/ACE-Brain-0-8B
```

ACE-Brain commited on Feb 26

Commit

e9036d3

verified ·

1 Parent(s): f756d24

Upload README.md

Browse files

Files changed (1) hide show

README.md +46 -12

README.md CHANGED Viewed

@@ -38,7 +38,6 @@ Extensive evaluation across **24** benchmarks demonstrates that ACE-Brain achiev
 ## Key Features
 - Unified multimodal foundation model for embodied intelligence
 - Strong spatial reasoning as a universal intelligence scaffold
 - Supports diverse embodiment platforms:
@@ -49,18 +48,8 @@ Extensive evaluation across **24** benchmarks demonstrates that ACE-Brain achiev
 - Cross-domain generalization across perception, reasoning, and planning
 - Evaluated on 24 real-world embodied intelligence benchmarks
-## Core Capabilities
-<div align="center">
-  <img src="./assets/fig2.png" width=800>
-</div>
 ## Performance Highlights
-<div align="center">
-  <img src="./assets/radarchart.png" width=800>
-</div>
 ACE-Brain achieves strong performance across **24 benchmarks covering Spatial Intelligence, Embodied Interaction, Autonomous Driving, and Low-Altitude Sensing**, consistently outperforming existing open-source embodied VLMs and remaining competitive with closed-source models.
 The model shows robust capability in **spatial reasoning, physical interaction understanding, task-oriented decision-making, and dynamic scene interpretation**, enabling reliable performance across diverse real-world embodiment scenarios.
@@ -96,9 +85,54 @@ Despite its domain specialization, ACE-Brain maintains strong general multimodal
   <img src="./assets/table4.png" width=800>
 </div>
-> **Bold** numbers indicate the best results, <u>underlined</u> numbers indicate the second-best results, and results marked with \* are obtained using our evaluation framework.
 ## Citation

 ## Key Features
 - Unified multimodal foundation model for embodied intelligence
 - Strong spatial reasoning as a universal intelligence scaffold
 - Supports diverse embodiment platforms:
 - Cross-domain generalization across perception, reasoning, and planning
 - Evaluated on 24 real-world embodied intelligence benchmarks
 ## Performance Highlights
 ACE-Brain achieves strong performance across **24 benchmarks covering Spatial Intelligence, Embodied Interaction, Autonomous Driving, and Low-Altitude Sensing**, consistently outperforming existing open-source embodied VLMs and remaining competitive with closed-source models.
 The model shows robust capability in **spatial reasoning, physical interaction understanding, task-oriented decision-making, and dynamic scene interpretation**, enabling reliable performance across diverse real-world embodiment scenarios.
   <img src="./assets/table4.png" width=800>
 </div>
+> **Bold** numbers indicate the best results, <u>underlined</u> numbers indicate the second-best results, and results marked with \* are obtained using our evaluation framework.
+## Inference Example
+```
+from transformers import Qwen3VLForConditionalGeneration, AutoProcessor
+# default: Load the model on the available device(s)
+model = Qwen3VLForConditionalGeneration.from_pretrained(
+    "ACE-Brain/ACE-Brain-8B", dtype="auto", device_map="auto"
+)
+processor = AutoProcessor.from_pretrained("ACE-Brain/ACE-Brain-8B")
+messages = [
+    {
+        "role": "user",
+        "content": [
+            {
+                "type": "image",
+                "image": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg",
+            },
+            {"type": "text", "text": "Describe this image."},
+        ],
+    }
+]
+# Preparation for inference
+inputs = processor.apply_chat_template(
+    messages,
+    tokenize=True,
+    add_generation_prompt=True,
+    return_dict=True,
+    return_tensors="pt"
+)
+inputs = inputs.to(model.device)
+# Inference: Generation of the output
+generated_ids = model.generate(**inputs, max_new_tokens=128)
+generated_ids_trimmed = [
+    out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
+]
+output_text = processor.batch_decode(
+    generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
+)
+print(output_text)
+```
 ## Citation