mix_77/gemma3-qat-tools:4b

mix_77/ gemma3-qat-tools:4b

1,801 Downloads Updated 11 months ago

The current, most capable model that runs on a single GPU. With quantization and tools.

vision tools 4b 12b 27b

ollama run mix_77/gemma3-qat-tools:4b

curl http://localhost:11434/api/chat \
  -d '{
    "model": "mix_77/gemma3-qat-tools:4b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='mix_77/gemma3-qat-tools:4b',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'mix_77/gemma3-qat-tools:4b',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 11 months ago

11 months ago

36fa44ea4aa4 · 4.0GB ·

model

archgemma3

parameters4.3B

quantizationQ4_0

4.0GB

template

{{- if .Messages }} {{- if or .System .Tools }}<start_of_turn>user {{- if .System}} {{ .System }} {{

1.5kB

params

{ "stop": [ "<end_of_turn>" ], "temperature": 1, "top_k": 64, "top_p": 0

77B

Readme

Gemma3 with quantization and tool support created based on the 4b-it-qat, 12b-it-qat and 27b-it-qat models of the official gemma3 image (all the original licenses apply)

Tools prompt is written according to this tutorial