WhiteCodex/LFM2.5-THINKING-FINETUNE-V7

TEXT GENERATIONConcurrency Cost:1Model Size:1.2BQuant:BF16Ctx Length:32kPublished:Apr 27, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

WhiteCodex/LFM2.5-THINKING-FINETUNE-V7 is a 1.2 billion parameter language model developed by WhiteCodex, fine-tuned from LFM2.5-THINKING-FINETUNE-V6. This model was trained with a 32768 token context length, utilizing Unsloth and Huggingface's TRL library for accelerated training. It is designed for general language understanding and generation tasks, building upon its predecessor's capabilities.

Loading preview...

Model Overview

WhiteCodex/LFM2.5-THINKING-FINETUNE-V7 is a 1.2 billion parameter language model developed by WhiteCodex, representing an iteration fine-tuned from the LFM2.5-THINKING-FINETUNE-V6 model. This version incorporates training optimizations that resulted in a 2x speed improvement.

Key Characteristics

  • Parameter Count: 1.2 billion parameters, offering a balance between performance and computational efficiency.
  • Context Length: Supports a substantial context window of 32768 tokens, enabling processing of longer inputs and generating more coherent, extended outputs.
  • Training Efficiency: Leverages Unsloth and Huggingface's TRL library, which contributed to a 2x faster training process compared to previous iterations.
  • Development: Developed by WhiteCodex, indicating a continued focus on refining their LFM2.5 series.
  • License: Released under the Apache-2.0 license, allowing for broad use and distribution.

Potential Use Cases

Given its parameter count and context length, this model is suitable for a variety of natural language processing tasks, including:

  • Text generation and completion.
  • Summarization of longer documents.
  • Question answering over extensive texts.
  • General conversational AI applications where context retention is important.

Finetunes
1 models