Downloads · 30 days
33
22% of all-time downloads
distil-lfm25-voice-assistant
distil-labs/distil-lfm25-voice-assistant
distil-lfm25-voice-assistant is a text generation model from distil-labs. Use it when you need the model to write or continue text. The card lists the license as other.
A fine-tuned version of LiquidAI/LFM2.5-350M for multi-turn banking voice assistant via tool calling, trained using the distil labs platform.
All-time downloads
151
Public
Parameters
354M
709 MB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors709 MB · 99%
At a glance
- Task
- Text Generation
- License
- other
- Model type
- lfm2
- Access
- Public
- Created
- Mar 30, 2026
- Updated
- Mar 30, 2026
- SHA
- 838468d2
Try a prompt
Base models
Model card
From the Hugging Face model README
distil-lfm25-voice-assistant
A fine-tuned version of LiquidAI/LFM2.5-350M for multi-turn banking voice assistant via tool calling, trained using the distil labs platform.
This model routes customer requests to the correct banking function (check balance, transfer money, cancel card, report fraud, etc.) while extracting required parameters. With 14 distinct functions, complex slot types, and ASR transcription artifacts in the input, this is the hardest task in our benchmark suite.
Results
| Metric | Teacher (120B) | LFM2.5-350M Base | LFM2.5-350M Tuned |
|---|---|---|---|
| Tool Call Equivalence | 96.95% | 34.5% | 95.9% |
| ROUGE | 97.55% | 74.1% | 99.0% |
The tuned model comes within 1.1 points of the 120B teacher on the hardest task in our benchmark suite.
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-voice-assistant-banking |
| Training method | SFT with LoRA |
| Platform | distil labs |
Training Progress
| Epoch | Tool Call Equivalence |
|---|---|
| 0 (base) | 34.5% |
| 1 | 86.8% |
| 2 | 92.4% |
| 3 | 95.9% |
| 4 | 95.4% |
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 targets on-device banking workflows where customer data cannot leave the device perimeter, deployable on mobile NPUs via ONNX or on Apple Silicon laptops via MLX for branch-level kiosk applications. 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.
- Task
- Text Generation
- Type
- lfm2
- License
- other
- Created
- Mar 30, 2026
- Updated
- Mar 30, 2026