Instructions to use LiquidAI/LFM2-1.2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LiquidAI/LFM2-1.2B with Transformers:

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

pipe = pipeline("text-generation", model="LiquidAI/LFM2-1.2B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LiquidAI/LFM2-1.2B")
model = AutoModelForCausalLM.from_pretrained("LiquidAI/LFM2-1.2B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use LiquidAI/LFM2-1.2B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LiquidAI/LFM2-1.2B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LiquidAI/LFM2-1.2B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LiquidAI/LFM2-1.2B

SGLang

How to use LiquidAI/LFM2-1.2B 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 "LiquidAI/LFM2-1.2B" \
    --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": "LiquidAI/LFM2-1.2B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "LiquidAI/LFM2-1.2B" \
        --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": "LiquidAI/LFM2-1.2B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use LiquidAI/LFM2-1.2B with Docker Model Runner:
```
docker model run hf.co/LiquidAI/LFM2-1.2B
```

Support tool calls

by CISCai - opened Jul 12, 2025

base: refs/heads/main

←

from: refs/pr/6

Discussion Files changed

+49

-4

CISCai

Jul 12, 2025

No description provided.

Support tool calls320fdd0d

mlabonne

Liquid AI org Jul 15, 2025

Hey, thanks for your PR! We need some time to craft a consolidated chat template, but we'll take your suggestion into account.

TroyDoesAI

Aug 9, 2025

•

edited Aug 9, 2025

All love friends, I finally got around to applying my research on your model for interruptible streams for parallel and sequential tool calling and taught it to do some light reasoning on top of your pirate version @mlabonne and it absolutely crushes tool calling more consistent than many much larger models. Nice job guys.

mlabonne

Liquid AI org Aug 12, 2025

Thanks for the video @TroyDoesAI , it looks great! :)

Tonic

Aug 15, 2025

All love friends, I finally got around to applying my research on your model for interruptible streams for parallel and sequential tool calling and taught it to do some light reasoning on top of your pirate version @mlabonne and it absolutely crushes tool calling more consistent than many much larger models. Nice job guys.

this is super cool , would love to take a peak at the pirate code if it's opened somewhere

mlabonne

Liquid AI org Aug 15, 2025

Sure, the pirate model was used with the Axolotl SFT notebook.

TroyDoesAI

Aug 29, 2025

A little update for my experiments with Liquid calling a larger model function calling is pretty cool. https://youtube.com/shorts/zgrxHFKesPI?si=ii2X2gftNfUmJEkO

Ghadir12

Aug 31, 2025

hello

mlabonne

Liquid AI org Sep 1, 2025

Thanks @TroyDoesAI , very nice demo! Enjoy your cake :)

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

Cannot merge

This branch has merge conflicts in the following files:

chat_template.jinja

· Sign up or log in to comment