Instructions to use ngqtrung/M3OPD-Reasoner-9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ngqtrung/M3OPD-Reasoner-9B with Transformers:

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

pipe = pipeline("image-text-to-text", model="ngqtrung/M3OPD-Reasoner-9B")
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("ngqtrung/M3OPD-Reasoner-9B")
model = AutoModelForMultimodalLM.from_pretrained("ngqtrung/M3OPD-Reasoner-9B", 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 ngqtrung/M3OPD-Reasoner-9B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ngqtrung/M3OPD-Reasoner-9B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ngqtrung/M3OPD-Reasoner-9B",
		"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/ngqtrung/M3OPD-Reasoner-9B

SGLang

How to use ngqtrung/M3OPD-Reasoner-9B 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 "ngqtrung/M3OPD-Reasoner-9B" \
    --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": "ngqtrung/M3OPD-Reasoner-9B",
		"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 "ngqtrung/M3OPD-Reasoner-9B" \
        --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": "ngqtrung/M3OPD-Reasoner-9B",
		"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 ngqtrung/M3OPD-Reasoner-9B with Docker Model Runner:
```
docker model run hf.co/ngqtrung/M3OPD-Reasoner-9B
```

M3OPD-Reasoner-9B

Main model of the M3OPD-Reasoner recipe. It starts from the final MOPD student (M3OPD-MOPD-9B) and continues it with MixRL: GSPO on all prompts of M3OPD-RL-193K, with the reward and the settings of the domain teachers and of the MixRL baseline. The recipe has three stages: domain teachers (SFT warm-up, then GSPO), MOPD of the three teachers into one student, and MixRL on top of the student. This is step 180 of the MixRL stage.

All checkpoints of the recipe are in the M3OPD-Reasoner collection. The data are M3OPD-SFT-1.35M and M3OPD-RL-193K.

Usage

The model has the architecture, tokenizer, processor and chat template of Qwen3.5-9B, so it runs wherever Qwen3.5-9B runs (Transformers, vLLM, SGLang). We train and evaluate with the native Qwen3.5 chat template, thinking enabled, no system prompt, and one answer-format line after the question, for example Put your final answer within \boxed{}.

vllm serve ngqtrung/M3OPD-Reasoner-9B --max-model-len 40960

License

Apache 2.0, the license of Qwen3.5-9B.

Downloads last month: 22

Safetensors

Model size

9B params

Tensor type

BF16

Model tree for ngqtrung/M3OPD-Reasoner-9B

Base model

Qwen/Qwen3.5-9B-Base

Finetuned

Qwen/Qwen3.5-9B

Finetuned

(1,305)

this model

Datasets used to train ngqtrung/M3OPD-Reasoner-9B

Collection including ngqtrung/M3OPD-Reasoner-9B

M3OPD-Reasoner

Collection

Data and checkpoints of M3OPD-Reasoner, an open recipe for reasoning vision-language models with multi-teacher on-policy distillation. • 8 items • Updated 2 days ago