mannix/llama3-12b

mannix/ llama3-12b:latest

464 Downloads Updated 1 year ago

Meta-Llama-3-12B-Instruct is a depth upscaling merge of llama3-8b from M. Labonne

ollama run mannix/llama3-12b

curl http://localhost:11434/api/chat \
  -d '{
    "model": "mannix/llama3-12b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='mannix/llama3-12b',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'mannix/llama3-12b',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 2 years ago

2 years ago

e0c119d95b71 · 6.6GB ·

model

archllama

parameters11.5B

quantizationQ4_0

6.6GB

params

{ "num_keep": 24, "stop": [ "<|start_header_id|>", "<|end_header_id|>",

110B

template

{{ if .System }}<|start_header_id|>system<|end_header_id|> {{ .System }}<|eot_id|>{{ end }}{{ if .Pr

260B

Readme

Meta-Llama-3-12B-Instruct is a depth upscaling merge of llama3-8b from M. Labonne

Quantized with imatrix using groups_merged.txt

https://huggingface.co/mlabonne/Meta-Llama-3-12B-Instruct