tinyllama:1.1b-chat-v0.6-q4

tinyllama:1.1b-chat-v0.6-q4_1

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:1.1b-chat-v0.6-q4_1

curl http://localhost:11434/api/chat \
  -d '{
    "model": "tinyllama:1.1b-chat-v0.6-q4_1",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='tinyllama:1.1b-chat-v0.6-q4_1',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'tinyllama:1.1b-chat-v0.6-q4_1',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 2 years ago

2 years ago

ad5344d40d22 · 702MB ·

model

archllama

parameters1.1B

quantizationQ4_1

702MB

params

{ "stop": [ "<|system|>", "<|user|>", "<|assistant|>", "</s>"

98B

system

You are a helpful AI assistant.

31B

template

<|system|> {{ .System }}</s> <|user|> {{ .Prompt }}</s> <|assistant|>

70B

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