Vortex5/Moonlit-Shadow-12B

TEXT GENERATIONConcurrency Cost:1Model Size:12BQuant:FP8Ctx Length:32kTool Calling:SupportedPublished:Aug 18, 2025Architecture:Transformer0.0K Cold

Vortex5/Moonlit-Shadow-12B is a 12 billion parameter language model created by Vortex5, leveraging the Mistral-Nemo-Instruct-2407 base model. This model was developed using the Model Stock merge method, combining eleven distinct pre-trained models to enhance its capabilities. It is designed for general language tasks, benefiting from the diverse strengths of its constituent models and supporting a 32768 token context length.

Loading preview...

Overview

Vortex5/Moonlit-Shadow-12B is a 12 billion parameter language model built upon the mistralai/Mistral-Nemo-Instruct-2407 base. It was created using the Model Stock merge method, which combines multiple pre-trained models to achieve a more robust and versatile output. This approach integrates diverse linguistic and reasoning capabilities from its constituent models, aiming for broad applicability in various language understanding and generation tasks.

Key Capabilities

  • Merged Architecture: Utilizes the Model Stock method to blend the strengths of eleven different 12B parameter models, including nothingiisreal/MN-12B-Celeste-V1.9, inflatebot/MN-12B-Mag-Mell-R1, and anthracite-org/magnum-v4-12b, among others.
  • Base Model Foundation: Benefits from the strong performance characteristics of the Mistral-Nemo-Instruct-2407 base model.
  • Context Length: Supports a substantial context window of 32768 tokens, enabling processing of longer inputs and generating more coherent, extended responses.

Good For

  • General-purpose text generation: Suitable for a wide array of tasks due to its merged nature.
  • Applications requiring diverse knowledge: The combination of multiple models suggests a broad knowledge base.
  • Experiments with merged models: Provides a practical example of the Model Stock merging technique.