Model Library

Liquid Foundation Models (LFMs) are a new class of multimodal architectures built for fast inference and on-device deployment. Browse all available models and formats here.

All of our models share the following capabilities:

• 32k token context length for extended conversations and document processing
• Designed for fast inference with Transformers, llama.cpp, vLLM, MLX, Ollama, and LEAP
• Trainable via SFT, DPO, and GRPO with LEAP Finetune, TRL, and Unsloth

Model Families

Choose a model based on your desired functionalities. Each individual model card has specific details on deployment and customization.

Text Models

Chat, tool calling, structured output, and classification.

Vision Models

Image understanding with LFM backbones and custom encoders.

Audio Models

Interleaved audio/text models for TTS, ASR, and voice chat.

Liquid Nanos

Task-specific models for extraction, summarization, RAG, and translation.

Model Formats

All LFM2 models are available in multiple formats for flexible deployment:

• GGUF — Best for local CPU/GPU inference on any platform. Use with llama.cpp, LM Studio, or Ollama. Append -GGUF to any model name.
• MLX — Best for Mac users with Apple Silicon. Leverages unified memory for fast inference via MLX. Browse at mlx-community.
• ONNX — Best for production deployments and edge devices. Cross-platform with ONNX Runtime across CPUs, GPUs, and accelerators. Append -ONNX to any model name.

Quantization

Quantization reduces model size and speeds up inference with minimal quality loss. Available options by format:

• GGUF — Supports Q2_K, Q3_K_M, Q4_K_M, Q5_K_M, Q6_K, and Q8_0 quantization levels. Q4_K_M offers the best balance of size and quality. Q8_0 preserves near-full precision.
• MLX — Available in 4bit and 8bit variants. 8bit is the default for most models.
• ONNX — Supports FP16 and INT8 quantization. INT8 is best for CPU inference; FP16 for GPU acceleration.

Model Chart

Model	HF	GGUF	MLX	ONNX	Trainable?
Text-to-text Models
LFM2.5 Models (Latest Release)
LFM2.5-1.2B-Instruct	✓	✓	✓	✓	Yes (TRL)
LFM2.5-1.2B-Thinking	✓	✓	✓	✓	Yes (TRL)
LFM2.5-1.2B-Base	✓	✓	✗	✓	Yes (TRL)
LFM2.5-1.2B-JP	✓	✓	✓	✓	Yes (TRL)
LFM2 Models
LFM2-8B-A1B	✓	✓	✓	✗	Yes (TRL)
LFM2-2.6B	✓	✓	✓	✓	Yes (TRL)
LFM2-1.2B Deprecated	✓	✓	✓	✓	Yes (TRL)
LFM2-700M	✓	✓	✓	✓	Yes (TRL)
LFM2-350M	✓	✓	✓	✓	Yes (TRL)
Vision Language Models
LFM2.5 Models (Latest Release)
LFM2.5-VL-1.6B	✓	✓	✓	✓	Yes (TRL)
LFM2 Models
LFM2-VL-3B	✓	✓	✓	✗	Yes (TRL)
LFM2-VL-1.6B	✓	✓	✓	✗	Yes (TRL)
LFM2-VL-450M	✓	✓	✓	✗	Yes (TRL)
Audio Models
LFM2.5 Models (Latest Release)
LFM2.5-Audio-1.5B	✓	✓	✗	✗	Yes (TRL)
LFM2 Models
LFM2-Audio-1.5B	✓	✓	✗	✗	No
Liquid Nanos
LFM2-1.2B-Extract	✓	✓	✗	✓	Yes (TRL)
LFM2-350M-Extract	✓	✓	✗	✓	Yes (TRL)
LFM2-350M-ENJP-MT	✓	✓	✓	✓	Yes (TRL)
LFM2-1.2B-RAG	✓	✓	✗	✓	Yes (TRL)
LFM2-1.2B-Tool Deprecated	✓	✓	✗	✓	Yes (TRL)
LFM2-350M-Math	✓	✓	✗	✓	Yes (TRL)
LFM2-350M-PII-Extract-JP	✓	✓	✗	✗	Yes (TRL)
LFM2-ColBERT-350M	✓	✗	✗	✗	Yes (PyLate)
LFM2-2.6B-Transcript	✓	✓	✗	✓	Yes (TRL)