Instructions to use ATH-MaaS/Ovis2-16B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ATH-MaaS/Ovis2-16B with Transformers:

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

pipe = pipeline("image-text-to-text", model="ATH-MaaS/Ovis2-16B", 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 AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("ATH-MaaS/Ovis2-16B", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ATH-MaaS/Ovis2-16B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ATH-MaaS/Ovis2-16B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ATH-MaaS/Ovis2-16B",
		"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/ATH-MaaS/Ovis2-16B

SGLang

How to use ATH-MaaS/Ovis2-16B 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 "ATH-MaaS/Ovis2-16B" \
    --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": "ATH-MaaS/Ovis2-16B",
		"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 "ATH-MaaS/Ovis2-16B" \
        --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": "ATH-MaaS/Ovis2-16B",
		"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 ATH-MaaS/Ovis2-16B with Docker Model Runner:
```
docker model run hf.co/ATH-MaaS/Ovis2-16B
```

xxyyy123 commited on Feb 11, 2025

Commit

f990da2

verified ·

1 Parent(s): 5090605

Update modeling_ovis.py

Browse files

Files changed (1) hide show

modeling_ovis.py +4 -4

modeling_ovis.py CHANGED Viewed

@@ -288,10 +288,10 @@ class Ovis(OvisPreTrainedModel):
         super().__init__(config, *inputs, **kwargs)
         attn_kwargs = dict()
         if self.config.llm_attn_implementation:
-            if self.config.llm_attn_implementation == "flash_attention_2":
-                assert (is_flash_attn_2_available() and
-                        version.parse(importlib.metadata.version("flash_attn")) >= version.parse("2.6.3")), \
-                    "Using `flash_attention_2` requires having `flash_attn>=2.6.3` installed."
             attn_kwargs["attn_implementation"] = self.config.llm_attn_implementation
         self.llm = AutoModelForCausalLM.from_config(self.config.llm_config, **attn_kwargs)
         assert self.config.hidden_size == self.llm.config.hidden_size, "hidden size mismatch"

         super().__init__(config, *inputs, **kwargs)
         attn_kwargs = dict()
         if self.config.llm_attn_implementation:
+            # if self.config.llm_attn_implementation == "flash_attention_2":
+            #     assert (is_flash_attn_2_available() and
+            #             version.parse(importlib.metadata.version("flash_attn")) >= version.parse("2.6.3")), \
+            #         "Using `flash_attention_2` requires having `flash_attn>=2.6.3` installed."
             attn_kwargs["attn_implementation"] = self.config.llm_attn_implementation
         self.llm = AutoModelForCausalLM.from_config(self.config.llm_config, **attn_kwargs)
         assert self.config.hidden_size == self.llm.config.hidden_size, "hidden size mismatch"