bunnycore/CyberCore-Qwen-2.1-7B

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:7.6BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Nov 21, 2024Architecture:Transformer0.0K Featherless Exclusive Cold

bunnycore/CyberCore-Qwen-2.1-7B is a 7.6 billion parameter language model created by bunnycore through a TIES merge of several Qwen-based models, including fblgit/cybertron-v4-qw7B-UNAMGS and fblgit/cybertron-v4-qw7B-MGS, both combined with bunnycore/Qwen-2.1-7b-Persona-lora_model. Built upon rombodawg/Rombos-LLM-V2.5-Qwen-7b as its base, this model leverages a 32768 token context length. Its architecture is designed to integrate diverse capabilities from its constituent models, making it suitable for applications requiring a blend of their respective strengths.

Loading preview...

Overview

bunnycore/CyberCore-Qwen-2.1-7B is a 7.6 billion parameter language model developed by bunnycore. It was created using the TIES merge method via mergekit, combining multiple specialized Qwen-based models. The base model for this merge is rombodawg/Rombos-LLM-V2.5-Qwen-7b, and it supports a context length of 32768 tokens.

Key Capabilities

  • Integrated Strengths: Merges capabilities from fblgit/cybertron-v4-qw7B-UNAMGS and fblgit/cybertron-v4-qw7B-MGS, each enhanced with bunnycore/Qwen-2.1-7b-Persona-lora_model.
  • Efficient Merging: Utilizes the TIES (Trimmed, Iterative, and Selective) merging technique to combine model weights effectively.
  • Qwen-2.1 Architecture: Benefits from the underlying Qwen-2.1 architecture, known for its strong performance across various language tasks.

Good for

  • Exploratory AI Development: Ideal for researchers and developers looking to experiment with models that combine diverse characteristics from multiple specialized components.
  • Applications Requiring Blended Expertise: Suitable for use cases where a single model needs to exhibit a mix of capabilities derived from its merged constituents, potentially offering a more generalized performance profile than its individual components.
  • Leveraging Qwen-2.1 Ecosystem: Users already familiar with or building upon the Qwen-2.1 family can easily integrate and utilize this merged model.

Finetunes
4 models