This model converts natural language commands ("turn off the kitchen lights," "set the thermostat to 72") into structured function calls, handling multi-turn conversations where users adjust commands or issue sequences.

Results

Metric	Teacher (120B)	LFM2.5-350M Base	LFM2.5-350M Tuned
Tool Call Equivalence	92.11%	63.2%	96.7%
ROUGE	98.53%	94.6%	99.4%

The tuned model beats the 120B teacher by 4.6 percentage points on smart home control.

Training Details

Parameter	Value
Base model	LiquidAI/LFM2.5-350M
Teacher model	GPT-oss-120B
Task type	Multi-turn tool calling (closed-book)
Training data	distil-labs/distil-smart-home
Training method	SFT with LoRA
Platform	distil labs

Training Progress

Epoch	Tool Call Equivalence
0 (base)	63.2%
1	96.1%
2	96.1%
3	96.7%
4	96.7%

Usage

This model uses the LFM2.5 tool calling format with <|tool_call_start|> and <|tool_call_end|> tags:

<|tool_call_start|>[function_name(arg1="value1", arg2=42)]<|tool_call_end|>

Deployment

This model is a natural fit for embedded deployment on smart home hubs and IoT gateways using the ONNX runtime, where sub-second latency on a low-power NPU means voice commands execute without a round trip to the cloud. It also works with Ollama, vLLM, llama.cpp, or any inference runtime that supports Safetensors.

Blog Post

For the full writeup, see: Fine-Tuning Liquid's LFM2.5: Accurate Tool Calling at 350M Parameters

License

This model is licensed under the LFM Open Model License v1.0.

distil-labs/distil-lfm25-home-assistant

distil-lfm25-home-assistant