TinyLlama/TinyLlama-1.1B-Chat-v0.5
Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:1.1BQuant:BF16Ctx Length:2kPublished:Nov 20, 2023License:apache-2.0Architecture:Transformer0.0K Open Weights Warm

TinyLlama/TinyLlama-1.1B-Chat-v0.5 is a 1.1 billion parameter Llama 2-based conversational language model developed by the TinyLlama project. It is fine-tuned for chat applications using the OpenAssistant/oasst_top1_2023-08-25 dataset. This compact model is designed for applications requiring restricted computation and memory footprints, offering Llama 2 compatibility in a smaller package.

Loading preview...

TinyLlama-1.1B-Chat-v0.5 Overview

TinyLlama-1.1B-Chat-v0.5 is a compact, 1.1 billion parameter language model developed by the TinyLlama project. It adopts the exact same architecture and tokenizer as Llama 2, ensuring broad compatibility with existing open-source projects built on Llama. The base model was pretrained on 3 trillion tokens, and this specific version is a chat model fine-tuned on the OpenAssistant/oasst_top1_2023-08-25 dataset, following the ChatML format.

Key Capabilities

  • Llama 2 Compatibility: Shares architecture and tokenizer with Llama 2, allowing for seamless integration into Llama-based workflows.
  • Compact Size: With only 1.1 billion parameters, it is optimized for environments with limited computational resources and memory.
  • Conversational AI: Fine-tuned specifically for chat-based interactions, making it suitable for dialogue systems.

Good For

  • Edge Devices: Deploying conversational AI on hardware with strict memory and processing constraints.
  • Rapid Prototyping: Quickly developing and testing chat applications where a full-sized Llama model might be overkill.
  • Educational Purposes: Understanding Llama 2 architecture and fine-tuning processes with a smaller, more manageable model.