Instructions to use rahul7star/LFM2.5-VL-3B-tool-sft-merged with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use rahul7star/LFM2.5-VL-3B-tool-sft-merged with Transformers:

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

pipe = pipeline("image-text-to-text", model="rahul7star/LFM2.5-VL-3B-tool-sft-merged", 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)

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

processor = AutoProcessor.from_pretrained("rahul7star/LFM2.5-VL-3B-tool-sft-merged", trust_remote_code=True)
model = AutoModelForMultimodalLM.from_pretrained("rahul7star/LFM2.5-VL-3B-tool-sft-merged", trust_remote_code=True, 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 rahul7star/LFM2.5-VL-3B-tool-sft-merged with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "rahul7star/LFM2.5-VL-3B-tool-sft-merged"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rahul7star/LFM2.5-VL-3B-tool-sft-merged",
		"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/rahul7star/LFM2.5-VL-3B-tool-sft-merged

SGLang

How to use rahul7star/LFM2.5-VL-3B-tool-sft-merged 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 "rahul7star/LFM2.5-VL-3B-tool-sft-merged" \
    --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": "rahul7star/LFM2.5-VL-3B-tool-sft-merged",
		"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 "rahul7star/LFM2.5-VL-3B-tool-sft-merged" \
        --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": "rahul7star/LFM2.5-VL-3B-tool-sft-merged",
		"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"
						}
					}
				]
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use rahul7star/LFM2.5-VL-3B-tool-sft-merged with Docker Model Runner:
```
docker model run hf.co/rahul7star/LFM2.5-VL-3B-tool-sft-merged
```

DEMO

rahul7star/LFM-3B-SFT-Demo

Run model

from transformers import pipeline
from IPython.display import display, Markdown
from PIL import Image
import requests
from io import BytesIO

# Load model
pipe = pipeline(
    "image-text-to-text",
    model="rahul7star/LFM2.5-VL-3B-tool-sft-merged",
    trust_remote_code=True,
)

# Conversation history
messages = []

def chat(user_text, image_url=None):
    global messages

    content = []

    if image_url:
        content.append({
            "type": "image",
            "url": image_url,
        })

    content.append({
        "type": "text",
        "text": user_text,
    })

    messages.append({
        "role": "user",
        "content": content,
    })

    output = pipe(text=messages)

    # Pipeline output is normally a list containing the generated conversation
    result = output[0]

    if isinstance(result, dict) and "generated_text" in result:
        generated = result["generated_text"]
    else:
        generated = result

    # If the pipeline returns the complete conversation,
    # extract the assistant's latest message.
    if isinstance(generated, list):
        assistant_message = generated[-1]

        if isinstance(assistant_message, dict):
            assistant_content = assistant_message.get("content", "")
        else:
            assistant_content = str(assistant_message)
    else:
        assistant_content = str(generated)

    messages.append({
        "role": "assistant",
        "content": [
            {
                "type": "text",
                "text": assistant_content,
            }
        ],
    })

    print("\nAssistant:")
    print(assistant_content)

    return assistant_content


# Test with image
image_url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"

chat(
    "What animal is on the candy?",
    image_url=image_url,
)