mingxilei/LFM2-1.2B-Tool-ft
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:1.2BQuant:BF16Context Size:32kPublished:Jan 21, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold
LFM2-1.2B-Tool-ft is a 1.2 billion parameter language model developed by mingxilei, fine-tuned from LiquidAI/LFM2-1.2B-Tool. This model was trained using Unsloth and Huggingface's TRL library, achieving 2x faster finetuning. It is designed for tool-use applications, leveraging its base model's capabilities for efficient integration with external functions and APIs.
Loading preview...
Overview
LFM2-1.2B-Tool-ft is a 1.2 billion parameter language model developed by mingxilei, specifically fine-tuned for tool-use applications. It builds upon the LiquidAI/LFM2-1.2B-Tool base model, enhancing its ability to interact with external tools and APIs.
Key Capabilities
- Efficient Finetuning: This model was finetuned using Unsloth and Huggingface's TRL library, resulting in a 2x speed improvement during the training process.
- Tool-Use Optimization: Designed to excel in scenarios requiring interaction with external functions, databases, or APIs, making it suitable for agentic workflows.
Good For
- Developing AI Agents: Ideal for creating agents that need to call external tools or functions to complete tasks.
- Automated Workflows: Suitable for integrating language capabilities into automated systems that interact with various software components.
- Resource-Efficient Deployment: With 1.2 billion parameters, it offers a balance between capability and computational efficiency, making it viable for deployment in environments with limited resources.