icefog72/IceDrunkenCherryRP-7b

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:7BQuant:FP8Context Size:4kTool Calling:SupportedPublished:Nov 21, 2024License:cc-by-nc-4.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

IceDrunkenCherryRP-7b by icefog72 is a 7 billion parameter language model, merged using the SLERP method from two previous Ice models. It is designed to handle a context window of 16k-25k tokens, potentially up to 32k, and is optimized for roleplay applications. The model is available in various quantized formats including Exl2 and GGUF.

Loading preview...

Overview

IceDrunkenCherryRP-7b is a 7 billion parameter language model developed by icefog72. This model was created using the SLERP merge method from two constituent models: icefog72/Ice0.29-06.11-RP and icefog72/Ice0.37-18.11-RP. It is noted for its ability to handle a substantial context window, estimated between 16,000 and 25,000 tokens, with potential support for up to 32,000 tokens.

Key Capabilities & Features

  • Roleplay Optimization: The model's naming convention and community engagement (Discord server for rules and examples) suggest a strong focus on roleplay scenarios.
  • Extended Context Window: Supports a large context, beneficial for maintaining coherence over long interactions.
  • Quantized Versions Available: Provided in various quantization formats, including:
  • Alpaca Format Compatibility: Generally works with the Alpaca instruction format.

Performance Metrics

Evaluated on the Open LLM Leaderboard, the model achieved an average score of 21.77. Specific scores include:

  • IFEval (0-Shot): 47.63
  • BBH (3-Shot): 31.51
  • MMLU-PRO (5-shot): 23.32

Good For

  • Applications requiring a model with a large context window.
  • Use cases focused on roleplay and interactive storytelling.
  • Developers looking for a 7B parameter model with optimized quantized versions for efficient deployment.

Finetunes
6 models