DanielClough/Candle_Mistral-7B-Instruct-v0.2

TEXT GENERATIONConcurrent Unit Cost:1Model Size:7BQuant:FP8Context Size:4kTool Calling:SupportedPublished:Feb 2, 2024License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

DanielClough/Candle_Mistral-7B-Instruct-v0.2 is a 7 billion parameter instruction-tuned causal language model, based on the Mistral-7B-Instruct-v0.2 architecture. This model is specifically packaged in the .gguf format for use with HuggingFace/Candle, making it suitable for efficient inference within the Candle framework. It maintains the original model's capabilities for general instruction following and text generation.

Loading preview...

Overview

DanielClough/Candle_Mistral-7B-Instruct-v0.2 is a specialized packaging of the Mistral-7B-Instruct-v0.2 model, designed for compatibility with the HuggingFace/Candle inference framework. This 7 billion parameter instruction-tuned model provides general language understanding and generation capabilities, inheriting the strengths of its base architecture.

Key Characteristics

  • Model Architecture: Based on the Mistral-7B-Instruct-v0.2, known for its strong performance in its size class.
  • Parameter Count: 7 billion parameters, offering a balance between performance and computational efficiency.
  • Context Length: Supports a context window of 4096 tokens.
  • Format: Provided in the .gguf format, specifically built for use with HuggingFace/Candle.
  • Inference Framework: Optimized for deployment and inference using the Candle library, which is a Rust-based deep learning framework.

Important Note

  • These .gguf files are not compatible with llama.cpp and are exclusively for use with HuggingFace/Candle. Users should refer to the original Mistral-7B-Instruct-v0.2 repository for comprehensive details regarding the model's training, capabilities, and ethical considerations.