mfielding92/gemini-3.1-pro-distill-reasoning-12B-QKVO-HF

VISIONConcurrency Cost:1Model Size:12BQuant:FP8Ctx Length:32kPublished:Feb 20, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

The mfielding92/gemini-3.1-pro-distill-reasoning-12B-QKVO-HF is a 12 billion parameter Gemma3 model, developed by mfielding92, fine-tuned for reasoning tasks. This model was trained using Unsloth and Huggingface's TRL library, achieving 2x faster training speeds. With a context length of 32768 tokens, it is optimized for applications requiring efficient and accelerated reasoning capabilities.

Loading preview...

Model Overview

The mfielding92/gemini-3.1-pro-distill-reasoning-12B-QKVO-HF is a 12 billion parameter Gemma3 model, developed by mfielding92. It is a fine-tuned variant of mfielding92/gemini-3.1-pro-distill-reasoning-12B-QVO-HF, specifically optimized for reasoning tasks.

Key Characteristics

  • Architecture: Based on the Gemma3 model family.
  • Parameter Count: 12 billion parameters.
  • Context Length: Supports a substantial context window of 32768 tokens.
  • Training Efficiency: This model was trained 2x faster using the Unsloth library in conjunction with Huggingface's TRL library, highlighting an efficient fine-tuning process.
  • License: Distributed under the Apache-2.0 license.

Use Cases

This model is particularly well-suited for applications that benefit from:

  • Reasoning Tasks: Its fine-tuning is geared towards enhancing reasoning capabilities.
  • Efficient Deployment: The accelerated training process suggests potential for more rapid iteration and deployment in development workflows.
  • Long Context Processing: The 32768 token context length makes it suitable for tasks requiring understanding and generation over extensive inputs.