MergekitCloud/mergekit-65

TEXT GENERATIONPricing:Input $0.37 / Cached $0.074 / Output $0.38Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Aug 28, 2026Architecture:Transformer Featherless Exclusive Cold

MergekitCloud/mergekit-65 is an 8 billion parameter language model created by merging several Llama 3.1-8B based models using the Model Stock method. This model leverages vicgalle/Humanish-Roleplay-Llama-3.1-8B as its base, integrating specialized fine-tunes for enhanced roleplay and conversational capabilities. It is designed to offer a versatile foundation for applications requiring nuanced and engaging interactive text generation.

Loading preview...

Overview

MergekitCloud/mergekit-65 is an 8 billion parameter language model developed by combining multiple specialized Llama 3.1-8B variants. This model was constructed using the Model Stock merge method, a technique detailed in the paper "Model Stock: A Method for Merging Pre-trained Language Models".

Key Capabilities

  • Enhanced Roleplay: Built upon a base of vicgalle/Humanish-Roleplay-Llama-3.1-8B, it integrates fine-tunes from models like ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.3, Undi95/Llama3-Unholy-8B-OAS, and Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2.
  • Merged Architecture: The Model Stock method allows for combining the strengths of several pre-trained models, potentially leading to a more robust and versatile output.
  • 8B Parameters: Offers a balance between performance and computational efficiency, suitable for various deployment scenarios.

Good For

  • Roleplaying Applications: Its foundational and merged components are specifically geared towards generating dynamic and engaging roleplay scenarios.
  • Conversational AI: The integration of models focused on human-like interaction suggests strong performance in dialogue systems.
  • Experimental Merging: Developers interested in exploring the outcomes of the Model Stock merging technique with Llama 3.1-8B derivatives.

Finetunes
1 models