LiquidAI/LFM2.5-1.2B-Instruct
LiquidAI/LFM2.5-1.2B-Instruct is a 1.17 billion parameter instruction-tuned model from the LFM2.5 family, developed by Liquid AI. It is designed for on-device deployment, offering best-in-class performance and fast edge inference with a 32,768 token context length. This model excels in agentic tasks, data extraction, and RAG, rivaling larger models while running efficiently on various hardware.
Loading preview...
LFM2.5-1.2B-Instruct: On-Device AI with Hybrid Architecture
LFM2.5-1.2B-Instruct is a 1.17 billion parameter instruction-tuned model developed by Liquid AI, part of the LFM2.5 family of hybrid models. It is specifically engineered for on-device deployment, offering high performance in a compact footprint.
Key Capabilities & Features
- Optimized for Edge Inference: Achieves fast decode speeds (e.g., 239 tok/s on AMD CPU, 82 tok/s on mobile NPU) and operates under 1GB of memory, with day-one support for llama.cpp, MLX, and vLLM.
- Scaled Training: Benefits from extended pre-training on 28 trillion tokens and large-scale multi-stage reinforcement learning, building upon the LFM2 architecture.
- Multilingual Support: Supports English, Arabic, Chinese, French, German, Japanese, Korean, and Spanish.
- Tool Use: Features robust function calling capabilities, allowing the model to interact with external tools for complex tasks.
- Long Context: Provides a substantial context length of 32,768 tokens.
- Performance: Demonstrates strong performance against other sub-2B models on benchmarks like GPQA (38.89), MMLU-Pro (44.35), and IFEval (86.23).
Use Cases & Recommendations
This model is particularly well-suited for:
- Agentic tasks
- Data extraction
- Retrieval Augmented Generation (RAG)
It is not recommended for knowledge-intensive tasks or programming. The model is available in various formats, including native, GGUF, ONNX, and MLX, to facilitate deployment across diverse hardware, from mobile devices to IoT systems.