DavidAU/LFM2.5-1.2B-Instruct-Thinking-Claude-High-Reasoning

TEXT GENERATIONConcurrency Cost:1Model Size:1.2BQuant:BF16Ctx Length:32kPublished:Jan 9, 2026Architecture:Transformer0.0K Cold

DavidAU/LFM2.5-1.2B-Instruct-Thinking-Claude-High-Reasoning is a 1.2 billion parameter instruction-tuned model by DavidAU, fine-tuned with Unsloth to integrate Claude Opus 4.5 High Reasoning capabilities. This model is optimized for generating explicit 'thinking/reasoning' text sections within its responses, rather than separate thought blocks, and supports a 32768 token context length. It excels at complex reasoning tasks, as evidenced by improved benchmark scores over its base model, making it suitable for applications requiring detailed internal thought processes.

Loading preview...

Overview

DavidAU/LFM2.5-1.2B-Instruct-Thinking-Claude-High-Reasoning is a 1.2 billion parameter model fine-tuned by DavidAU using Unsloth, specifically designed to emulate Claude Opus 4.5's high reasoning and thinking processes. It features a substantial 32768 token context window, enabling it to handle complex and lengthy prompts.

Key Capabilities

  • Explicit Reasoning: Generates detailed 'thinking/reasoning' sections directly within the response text, which can range from a few sentences to several paragraphs.
  • Enhanced Reasoning Performance: Shows improved benchmark scores across various tasks (e.g., ARC-Challenge, Hellaswag) compared to the standard LFM2.5 Instruct model.
  • Contextual Depth: Leverages a 32k context window to process and reason over extensive inputs.
  • Configurable Output: Users can influence the depth and presence of reasoning by using prompts like "Explain...", "Think deeply before you answer...", or "Plan out your reply...".

Good For

  • Complex Problem Solving: Ideal for use cases requiring the model to articulate its thought process before providing a final answer.
  • Interactive Reasoning: Suitable for applications where understanding the model's internal logic is beneficial, even if it requires regeneration to activate or expand reasoning sections.
  • Creative Writing & Story Generation: Demonstrated ability to plan and execute detailed narratives with psychological depth and atmospheric tension, as shown in the horror story and sci-fi examples.
  • Educational Tools: Can be used to illustrate reasoning steps for learning or analytical purposes.