Liquid AI

LFM 40B is a large language model from Liquid AI's Liquid Foundation Model (LFM) series. With 40 billion parameters, it is designed as a powerful general-purpose model with strong reasoning and instruction-following capabilities.

LFM2 1.2B is a lightweight, efficient language model from Liquid AI's LFM2 family. Designed for low-latency and resource-constrained environments, it provides foundational language capabilities in a compact 1.2 billion parameter package.

LFM2 2.6B is a compact and efficient language model from Liquid AI's LFM series. It is optimized for fast inference and low-cost deployment, making it suitable for edge devices and real-time applications.

LFM2 24B A2B is a 24-billion parameter language model from Liquid AI's second-generation Liquid Foundation Model (LFM2) series. It is designed as a general-purpose foundation model with strong reasoning and code generation capabilities.

LFM2 8B A1B is a language model from Liquid AI's LFM2 series. Based on its naming convention, it likely features an 8-billion parameter architecture with a specific variant (A1B), potentially optimized for efficiency or particular tasks. Further details on its capabilities require official documentation.

LFM2.5-1.2B-Instruct is a lightweight, efficient instruction-tuned model from Liquid AI's Liquid Foundation Model series. With only 1.2 billion parameters, it is optimized for fast inference and low-cost deployment, making it suitable for edge devices and applications requiring rapid response times.

LFM2.5-1.2B-Thinking is a compact, 1.2-billion parameter language model from Liquid AI, optimized for reasoning and thinking tasks. It is designed to deliver strong logical and analytical performance within a small, efficient footprint.

LFM2.5-VL-1.6B is a compact, 1.6-billion parameter vision-language model from Liquid AI's Liquid Foundation Model (LFM) series. It is designed for efficient multimodal understanding, processing both text and images, and is optimized for fast inference and low-cost deployment.