OrdenWills/LFM2.5-350M-home-assistant-dpo

TEXT GENERATIONConcurrency Cost:1Model Size:0.35BQuant:BF16Ctx Length:32kPublished:May 8, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

OrdenWills/LFM2.5-350M-home-assistant-dpo is a 350 million parameter language model developed by OrdenWills, fine-tuned from OrdenWills/LFM2.5-350M-home-assistant-sft-v13. This model was trained using Unsloth for accelerated performance, making it suitable for Home Assistant-related applications. It is designed for efficient deployment with its smaller parameter count and 32768 token context length.

Loading preview...

Model Overview

OrdenWills/LFM2.5-350M-home-assistant-dpo is a 350 million parameter language model developed by OrdenWills. It is a fine-tuned version of the OrdenWills/LFM2.5-350M-home-assistant-sft-v13 model, specifically optimized for Home Assistant applications.

Key Characteristics

  • Developer: OrdenWills
  • Base Model: Fine-tuned from OrdenWills/LFM2.5-350M-home-assistant-sft-v13.
  • Training Efficiency: This model was trained significantly faster using Unsloth, a library designed to accelerate large language model training.
  • License: Released under the Apache-2.0 license.

Use Cases

This model is particularly well-suited for:

  • Home Assistant Integrations: Its fine-tuning suggests strong performance in tasks related to the Home Assistant ecosystem.
  • Resource-Constrained Environments: With 350 million parameters, it offers a balance of capability and efficiency, making it suitable for deployment where computational resources are limited.
  • Applications requiring fast training: The use of Unsloth indicates a focus on rapid iteration and development cycles.