TinyLlama

Name: TinyLlama
Author: Singapore University of Technology and Design

Foundation model

TinyLlama

Release date: 2023-09-04Updated: 2024-01-06Views: 868

Live demoGitHub Hugging Face Compare

Parameters

1.1B

Context length

Multilingual

No data

Reasoning ability

No data

TinyLlama is an AI model published by Singapore University of Technology and Design, released on 2023-09-04, for Foundation model, with 1.1B parameters, and 2K context length, requiring about 4.4GB storage, with a 25.02 score on C-Eval.

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

TinyLlama

Model basics

Reasoning traces

No data

Thinking modes

Thinking modes not supported

Context length

2K tokens

Max output length

No data

Model type

Foundation model

Modality (in / out)

No data

Release date

2023-09-04

Model file size

4.4GB

MoE architecture

Total params / Active params

1.1B / Not applicable

Knowledge cutoff

No data

TinyLlama

Open source & experience

Code license

Apache 2.0

Weights license

Apache 2.0- Commercial use permitted

GitHub repo

github.com/jzhang38/TinyLlama

Hugging Face

huggingface.co/TinyLlama/TinyLlama-1.1B-interme…

Live demo

N/A

TinyLlama

Official resources

Paper

DataLearnerAI blog

N/A

TinyLlama

API details

API speed

No data

No public API pricing yet.

TinyLlama

Benchmark Results

TinyLlama currently shows benchmark results led by GSM8K (67 / 70, score 2.27), C-Eval (46 / 48, score 25.02), MBPP (68 / 70, score 19.91). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.