Instructions to use lzy-vlm-lab-opd/EAPO with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lzy-vlm-lab-opd/EAPO with Transformers:

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

pipe = pipeline("image-text-to-text", model="lzy-vlm-lab-opd/EAPO")
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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("lzy-vlm-lab-opd/EAPO")
model = AutoModelForMultimodalLM.from_pretrained("lzy-vlm-lab-opd/EAPO", device_map="auto")
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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use lzy-vlm-lab-opd/EAPO with vLLM:

Install from pip and serve model

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

SGLang

How to use lzy-vlm-lab-opd/EAPO 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 "lzy-vlm-lab-opd/EAPO" \
    --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": "lzy-vlm-lab-opd/EAPO",
		"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 "lzy-vlm-lab-opd/EAPO" \
        --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": "lzy-vlm-lab-opd/EAPO",
		"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 lzy-vlm-lab-opd/EAPO with Docker Model Runner:
```
docker model run hf.co/lzy-vlm-lab-opd/EAPO
```

RLSD + RLRT with failure-only EAPO Stage2, step 100

Run: hybrid-eapo-negative-log4-teacher100-pos05-stage1step150-full200-retry1. Stage2 global step: 100.

Starts from ZwZ Stage1 step150. Teacher mixture decays to zero over 100 steps; initial mixture 0.5, positive weight clip 0.5, negative weight clip 0.2.

Failed responses additionally use low-entropy EAPO credit allocation with kappa approximately log(4). Successful responses retain the RLRT branch. EAPO remains enabled after teacher influence reaches zero at step100. The mean baseline weight per failed response is preserved.

Load lzy-vlm-lab-opd/EAPO with revision="step100". This revision contains the student actor, processor, tokenizer and original generation configuration.

The export passed strict CPU loading, finite-parameter checks and a two-token multimodal generation smoke check. Benchmark evaluation has not been performed by this export workflow.

Downloads last month: 17

Safetensors

Model size

9B params

Tensor type

BF16

Model tree for lzy-vlm-lab-opd/EAPO

Base model

Qwen/Qwen3-VL-8B-Instruct

Finetuned

(683)

this model