TinyLlama-1.1B-Chat-v1.0 is a 1.1 billion parameter language model hosted on Hugging Face. It is provided as an open model with a defined chat template that structures conversations by inserting role-specific tags for user, system, and assistant messages, along with an end-of-sentence token.
The model includes an explicit chat template in its configuration that processes messages by prefixing content with tags such as <|user|, <|system|, and <|assistant|, appending the eos token after each, and adding a generation prompt for the final assistant turn when required. It specifies an eos token of </s, a pad token of </s, and an unk token of <unk, with use_default_system_prompt set to false. The repository records over 45 million all-time downloads and more than 2.4 million recent downloads.
It is delivered as a model card and associated files on the Hugging Face platform, where it can be accessed for inference through listed providers. The page was created on December 30, 2023.
In the Foundation models & chat space, TinyLlama 1.1B Chat takes a focused approach. It provides a lightweight, open-source conversational AI model for building chatbots and assistants. TinyLlama 1.1B Chat is an open-source project aimed at AI developers. The project is open source (Open Source). The product ships for the command line, and it can be self-hosted.
Behind TinyLlama 1.1B Chat is TinyLlama, and the product first shipped in 2023. Among its 5 catalogued features are Conversational AI, chat template support, and pretrained weights.
Latest indexed changes and source events
TinyLlama/TinyLlama-1.1B-Chat-v1.0 verified by the PulseGate indexer
