Instructions to use FlameF0X/LFM2.5-1.2B-Thinking-CodeX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use FlameF0X/LFM2.5-1.2B-Thinking-CodeX with Transformers:

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

pipe = pipeline("text-generation", model="FlameF0X/LFM2.5-1.2B-Thinking-CodeX")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("FlameF0X/LFM2.5-1.2B-Thinking-CodeX")
model = AutoModelForCausalLM.from_pretrained("FlameF0X/LFM2.5-1.2B-Thinking-CodeX")
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

vLLM

How to use FlameF0X/LFM2.5-1.2B-Thinking-CodeX with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "FlameF0X/LFM2.5-1.2B-Thinking-CodeX"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "FlameF0X/LFM2.5-1.2B-Thinking-CodeX",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/FlameF0X/LFM2.5-1.2B-Thinking-CodeX

SGLang

How to use FlameF0X/LFM2.5-1.2B-Thinking-CodeX 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 "FlameF0X/LFM2.5-1.2B-Thinking-CodeX" \
    --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": "FlameF0X/LFM2.5-1.2B-Thinking-CodeX",
		"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 "FlameF0X/LFM2.5-1.2B-Thinking-CodeX" \
        --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": "FlameF0X/LFM2.5-1.2B-Thinking-CodeX",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use FlameF0X/LFM2.5-1.2B-Thinking-CodeX with Docker Model Runner:
```
docker model run hf.co/FlameF0X/LFM2.5-1.2B-Thinking-CodeX
```

LFM2.5-1.2B-Thinking-CodeX / README.md

FlameF0X

Update README.md

f0f7945 verified about 7 hours ago

preview code

raw

history blame contribute delete

2.22 kB

metadata

base_model: LiquidAI/LFM2.5-1.2B-Thinking
library_name: transformers
model_name: LFM2.5-1.2B-Thinking-CodeX
tags:
  - generated_from_trainer
  - sft
  - trl
licence: license
datasets:
  - Modotte/CodeX-2M-Thinking
license: apache-2.0

LFM2.5-1.2B-Thinking-CodeX (Liquid CodeX)

LFM2.5-1.2B-Thinking-CodeX (Liquid CodeX) is a distillation of Claude into LFM2.5-1.2B-Thinking via LoRA.

Benchmark

Model	Average	HellaSwag	MMLU	Piqa	Source
FlameFOX/LFM2.5-1.2B-Distilled-Claude-4.6	46.76	39.51	31.99	68.77	Intel/low bit open llm leaderboard
FlameFOX/LFM2.5-1.2B-Thinking-CodeX	45.25	39.70	26.56	69.48	As the one from above

Quick start

from transformers import pipeline

question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="FlameF0X/LFM2.5-1.2B-Thinking-CodeX", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])

Training procedure

This model was trained with SFT.

Framework versions

TRL: 1.2.0
Transformers: 5.0.0
Pytorch: 2.10.0+cu128
Datasets: 4.8.4
Tokenizers: 0.22.2

Citations

Cite TRL as:

@software{vonwerra2020trl,
  title   = {{TRL: Transformers Reinforcement Learning}},
  author  = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
  license = {Apache-2.0},
  url     = {https://github.com/huggingface/trl},
  year    = {2020}
}