MergekitCloud/mergekit-10

TEXT GENERATIONPricing:Input $0.37 / Cached $0.074 / Output $0.38Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Aug 22, 2026Architecture:Transformer Featherless Exclusive Cold

MergekitCloud/mergekit-10 is an 8 billion parameter language model created by merging several Llama 3.1-8B based models using the Model Stock method, with vicgalle/Humanish-Roleplay-Llama-3.1-8B as its base. This model integrates capabilities from Undi95/Llama3-Unholy-8B-OAS, Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2, and ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.3. It is specifically designed to enhance roleplay and conversational generation, leveraging its merged components for nuanced and engaging interactions.

Loading preview...

Overview

MergekitCloud/mergekit-10 is an 8 billion parameter language model built upon the Llama 3.1-8B architecture. It was constructed using the Model Stock merge method, a technique described in the paper "Model Stock: A Method for Merging Pre-trained Language Models." The base model for this merge was vicgalle/Humanish-Roleplay-Llama-3.1-8B.

Key Capabilities

This model integrates the strengths of three distinct Llama 3.1-8B variants:

  • Undi95/Llama3-Unholy-8B-OAS: Contributes to its overall conversational and potentially unconstrained response generation.
  • Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2: Enhances its ability to generate diverse and less filtered content.
  • ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.3: Specifically designed to boost roleplay capabilities, providing more immersive and detailed character interactions.

Good For

Given its merged components, mergekit-10 is particularly well-suited for:

  • Advanced Roleplay Scenarios: Excelling in generating detailed and consistent character interactions.
  • Creative Writing: Producing imaginative and engaging narratives.
  • Open-ended Conversational Agents: Handling a wide range of prompts with nuanced responses.

This model leverages the combined knowledge and stylistic tendencies of its constituent models to offer a robust solution for applications requiring strong generative and interactive capabilities, especially in the realm of creative and roleplay-focused AI.