> ## Documentation Index
> Fetch the complete documentation index at: https://docs.liquid.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# FAQs

> Frequently asked questions about LFM models and deployment.

## General Questions

<Accordion title="What are LFM models?">
  LFM (Liquid Foundation Models) are a family of efficient language models built on a new hybrid architecture designed for fast training and inference. They range from 350M to 8B parameters and support text, vision, and audio modalities.
</Accordion>

<Accordion title="What context length do LFM models support?">
  All LFM models support a 32k token text context length for extended conversations and document processing.
</Accordion>

<Accordion title="Which inference frameworks are supported?">
  LFM models are compatible with:

  * [Transformers](/deployment/gpu-inference/transformers) - For research and development
  * [llama.cpp](/deployment/on-device/llama-cpp) - For efficient CPU inference
  * [vLLM](/deployment/gpu-inference/vllm) - For high-throughput production serving
  * [MLX](/deployment/on-device/mlx) - For Apple Silicon optimization
  * [Ollama](/deployment/on-device/ollama) - For easy local deployment
  * [LEAP](/deployment/on-device/sdk/quick-start) - For edge and mobile deployment
</Accordion>

## Model Selection

<Accordion title="Which model should I use for my use case?">
  * **General chat/instruction following**: LFM2.5-1.2B-Instruct (recommended)
  * **Vision tasks**: LFM2.5-VL-1.6B
  * **Audio/speech**: LFM2.5-Audio-1.5B
  * **Extraction tasks**: LFM2-1.2B-Extract or LFM2-350M-Extract
  * **Edge deployment**: LFM2-350M or LFM2-700M for smallest footprint
  * **Highest performance**: LFM2-8B-A1B (MoE architecture)
</Accordion>

<Accordion title="What is the difference between LFM2 and LFM2.5?">
  LFM2.5 models are updated versions with improved training that deliver higher performance while maintaining the same architecture. We recommend using LFM2.5 variants when available.
</Accordion>

<Accordion title="What are Liquid Nanos?">
  [Liquid Nanos](/lfm/models/liquid-nanos) are task-specific models fine-tuned for specialized use cases like:

  * Information extraction (LFM2-Extract)
  * Translation (LFM2-350M-ENJP-MT)
  * RAG question answering (LFM2-1.2B-RAG)
  * Meeting summarization (LFM2-2.6B-Transcript)
</Accordion>

## Deployment

<Accordion title="Can I run LFM models on mobile devices?">
  Yes! Use the [LEAP SDK](/deployment/on-device/sdk/quick-start) to deploy models on iOS and Android devices. LEAP provides optimized inference for edge deployment with support for quantized models.
</Accordion>

<Accordion title="What quantization formats are available?">
  * **GGUF**: For llama.cpp, LM Studio, Ollama (Q4\_0, Q4\_K\_M, Q5\_K\_M, Q6\_K, Q8\_0, F16)
  * **MLX**: For Apple Silicon (4-bit, 5-bit, 6-bit, 8-bit, bf16)
  * **ONNX**: For cross-platform deployment with ONNX Runtime
</Accordion>

<Accordion title="How do I choose the right quantization level?">
  * **Q4\_0 / 4-bit**: Smallest size, fastest inference, some quality loss
  * **Q8\_0 / 8-bit**: Good balance of size and quality
  * **F16 / bf16**: Full precision, best quality, largest size

  For most use cases, Q4\_K\_M or Q5\_K\_M provide good quality with significant size reduction.
</Accordion>

## Fine-tuning

<Accordion title="Can I fine-tune LFM models?">
  Yes! Most LFM models support fine-tuning with [TRL](/lfm/fine-tuning/trl) and [Unsloth](/lfm/fine-tuning/unsloth). Check the [Model Library](/lfm/models/complete-library) for trainability information.
</Accordion>

<Accordion title="What fine-tuning methods are supported?">
  * **LoRA/QLoRA**: Memory-efficient fine-tuning
  * **Full fine-tuning**: For maximum customization
  * **SFT (Supervised Fine-Tuning)**: For instruction tuning
</Accordion>

## Still Have Questions?

* Join our [Discord community](https://discord.gg/DFU3WQeaYD) for real-time help
* Check the [Cookbook](https://github.com/Liquid4All/cookbook) for examples
* See [Troubleshooting](/lfm/help/troubleshooting) for common issues