mindy-labs/mindy-7b-v2

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:8kPublished:Dec 14, 2023License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

Mindy-7b-v2 is a 7 billion parameter transformer-based language model developed by Mindy Group, Inc. This English-language model is a Frankenstein merge of AIDC-ai-business/Marcoroni-7B-v3 and Weyaxi/Seraph-7B, designed to combine their respective strengths. It is suitable for general language tasks where a 7B parameter model with an 8192 token context length is appropriate.

Loading preview...

Mindy-7b-v2 Overview

Mindy-7b-v2 is a 7 billion parameter, English-language, transformer-based language model developed by Mindy Group, Inc. It is notable for its unique construction as a "Frankenstein merge" of two distinct 7B models: AIDC-ai-business/Marcoroni-7B-v3 and Weyaxi/Seraph-7B. This merging approach aims to leverage the combined capabilities of its constituent models, offering a distinct profile compared to models trained from scratch or through standard fine-tuning.

Key Capabilities

  • Merged Architecture: Combines the strengths of Marcoroni-7B-v3 and Seraph-7B.
  • Language Support: Primarily focused on English language tasks.
  • Context Length: Supports an 8192 token context window, suitable for processing moderately long inputs.
  • General Purpose: Designed for a broad range of natural language processing applications.

Good For

  • Developers seeking a 7B parameter model with a unique merged lineage.
  • Applications requiring an English-centric language model with a reasonable context window.
  • Experimentation with merged model architectures for general text generation and understanding tasks.