tinyllama

3.5M Downloads Updated 2 years ago

The TinyLlama project is an open endeavor to train a compact 1.1B Llama model on 3 trillion tokens.

1.1b

ollama run tinyllama

curl http://localhost:11434/api/chat \
  -d '{
    "model": "tinyllama",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='tinyllama',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'tinyllama',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Models

View all →

Name

36 models

Size

Context

Input

tinyllama:latest

638MB · 2K context window · Text · 2 years ago

tinyllama:latest

638MB

Text

tinyllama:1.1b

latest

638MB · 2K context window · Text · 2 years ago

tinyllama:1.1b latest

638MB

Text

Readme

TinyLlama is a compact model with only 1.1B parameters. This compactness allows it to cater to a multitude of applications demanding a restricted computation and memory footprint.

References

Hugging Face

GitHub