Instructions to use aruntemme/LFM2.5-350M-M3-Distill with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use aruntemme/LFM2.5-350M-M3-Distill with Transformers:

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

pipe = pipeline("text-generation", model="aruntemme/LFM2.5-350M-M3-Distill")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("aruntemme/LFM2.5-350M-M3-Distill")
model = AutoModelForCausalLM.from_pretrained("aruntemme/LFM2.5-350M-M3-Distill", device_map="auto")
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

llama.cpp

How to use aruntemme/LFM2.5-350M-M3-Distill with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M

Use Docker

docker model run hf.co/aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M

LM Studio
Jan

vLLM

How to use aruntemme/LFM2.5-350M-M3-Distill with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "aruntemme/LFM2.5-350M-M3-Distill"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "aruntemme/LFM2.5-350M-M3-Distill",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M

SGLang

How to use aruntemme/LFM2.5-350M-M3-Distill 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 "aruntemme/LFM2.5-350M-M3-Distill" \
    --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": "aruntemme/LFM2.5-350M-M3-Distill",
		"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 "aruntemme/LFM2.5-350M-M3-Distill" \
        --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": "aruntemme/LFM2.5-350M-M3-Distill",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use aruntemme/LFM2.5-350M-M3-Distill with Ollama:
```
ollama run hf.co/aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M
```

Unsloth Studio

How to use aruntemme/LFM2.5-350M-M3-Distill with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for aruntemme/LFM2.5-350M-M3-Distill to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for aruntemme/LFM2.5-350M-M3-Distill to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for aruntemme/LFM2.5-350M-M3-Distill to start chatting

How to use aruntemme/LFM2.5-350M-M3-Distill with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use aruntemme/LFM2.5-350M-M3-Distill with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use aruntemme/LFM2.5-350M-M3-Distill with Docker Model Runner:
```
docker model run hf.co/aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M
```

Lemonade

How to use aruntemme/LFM2.5-350M-M3-Distill with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M

Run and chat with the model

lemonade run user.LFM2.5-350M-M3-Distill-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use aruntemme/LFM2.5-350M-M3-Distill with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default aruntemme/LFM2.5-350M-M3-Distill:Q4_K_M

Run Hermes

hermes

Atomic Chat

LFM2.5-350M-M3-Distill

A 350M-parameter assistant fine-tuned from LiquidAI/LFM2.5-350M on a synthetic instruction dataset distilled from MiniMax-M3 (a frontier teacher model). It targets fast, capable on-device / Raspberry Pi use for everyday assistant tasks — general chat, Q&A, summarization, text rewriting, classification/extraction, and JSON / structured output.

v2 (current). Same recipe as v1 but trained on a verbosity-curated dataset — terse tasks (rewrite, "make concise", tone changes) now give a single clean answer instead of multi-option essays, with no regression on chat / Q&A / summarization / JSON / tool-calling. Read Limitations before using.

What it's good at

Versus the stock LFM2.5-350M base (observed on a fixed 35-prompt eval set), the fine-tune produces:

More structured, complete answers for open-ended chat and explanations.
Better short-text classification (e.g. spam / sentiment) — fixed cases the base got wrong.
Cleaner summarization and JSON output for its size.
Tool / function calling is preserved from the base (native <|tool_call_start|> format) — and on math it correctly defers to a calculate tool instead of hallucinating arithmetic.

Intended use

General chat & advice
Factual Q&A
Summarization
Text rewriting (tone, formality, paraphrase, simple translation)
Classification & extraction (sentiment, spam, fields → JSON)
Structured / JSON output
Tool / function calling (native LFM2.5 format)

Limitations (please read)

Not for code or math. The LFM2.5 base is explicitly not recommended for code, math, or creative writing, and this model was deliberately not trained on the code/reasoning data. Use a code-capable model (e.g. Qwen3-0.6B) for those.
Weak at precise token-level editing. Grammar/spelling-correction tasks are unreliable — a genuine capacity limit at 350M (unchanged in v2).
Shallow synthesis. When summarizing long/multi-source material it tends to list findings rather than deeply synthesize — expected at this size.
Small model: it can hallucinate and should not be relied on for factual accuracy without verification.

Recommended generation settings

Use sampling (not greedy) — greedy decoding causes repetition loops at this size:

generation_config = dict(
    do_sample=True,
    temperature=0.7,
    top_p=0.9,
    repetition_penalty=1.1,
    max_new_tokens=512,
)

Usage (Transformers)

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

tok = AutoTokenizer.from_pretrained("aruntemme/LFM2.5-350M-M3-Distill")
model = AutoModelForCausalLM.from_pretrained("aruntemme/LFM2.5-350M-M3-Distill",
                                             dtype=torch.bfloat16).to("cuda")

msgs = [{"role": "user", "content": "Summarize the benefits of regular exercise in 3 bullet points."}]
inputs = tok.apply_chat_template(msgs, add_generation_prompt=True, return_tensors="pt",
                                 return_dict=True).to("cuda")
out = model.generate(**inputs, do_sample=True, temperature=0.7, top_p=0.9,
                     repetition_penalty=1.1, max_new_tokens=512)
print(tok.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))

Tool / function calling

Pass tools via the chat template; the model emits calls in LFM2.5's native format (<|tool_call_start|>[func(arg="val")]<|tool_call_end|>):

tools = [{"type": "function", "function": {
    "name": "get_weather", "description": "Get current weather for a city",
    "parameters": {"type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"]}}}]
msgs = [{"role": "user", "content": "What's the weather in Paris?"}]
inputs = tok.apply_chat_template(msgs, tools=tools, add_generation_prompt=True,
                                 return_tensors="pt", return_dict=True).to("cuda")
out = model.generate(**inputs, max_new_tokens=100)
# -> <|tool_call_start|>[get_weather(city="Paris")]<|tool_call_end|>

Usage (GGUF / llama.cpp / Raspberry Pi)

Quantized GGUF files are included for CPU / edge inference:

file	quant	approx size
`lfm2.5-350m-m3distill-Q4_K_M.gguf`	Q4_K_M	~230 MB
`lfm2.5-350m-m3distill-f16.gguf`	f16	~709 MB

./llama-cli -m lfm2.5-350m-m3distill-Q4_K_M.gguf \
  -p "Explain photosynthesis in 40 words." \
  --temp 0.7 --top-p 0.9 --repeat-penalty 1.1

Training

Base: LiquidAI/LFM2.5-350M
Teacher: MiniMax-M3 (synthetic data generated via the teacher, then quality-filtered/judged)
Method: full supervised fine-tuning (no LoRA), bf16, gradient checkpointing
Data (v2): ~30,400 instruction→response examples (chat, Q&A, summarization, rewriting, explanation, classification/extraction, JSON) — code & reasoning excluded to match the base's strengths; v2 additionally drops the multi-option / over-long responses on terse tasks to cut needless verbosity (v1 used ~32,100 uncurated examples)
Hyperparameters: 2 epochs, effective batch 32 (bs 4 × grad-accum 8), lr 1.5e-5 (cosine, 3% warmup), max sequence length 1024, weight decay 0.01
Hardware: single NVIDIA RTX 3060 (12 GB), ~2.7 h
Final eval: loss 1.87, token accuracy 0.60
Frameworks: TRL 1.6.0, Transformers 5.12.1, PyTorch 2.12.1

License & attribution

This is a derivative of LiquidAI/LFM2.5-350M, distributed under the LFM Open License v1.0 (commercial use permitted under a $10M annual-revenue threshold). This derivative is provided under the same license; a copy is included. The base model and its license notices are retained per the license terms.

Acknowledgements

Base model: Liquid AI (LFM2.5-350M)
Distillation teacher: MiniMax-M3

Downloads last month: 35

Safetensors

Model size

0.4B params

Tensor type

BF16

Model tree for aruntemme/LFM2.5-350M-M3-Distill

Base model

LiquidAI/LFM2.5-350M-Base

Finetuned

LiquidAI/LFM2.5-350M

Quantized

(52)

this model