aliafshar/gemma3-it-qat-tools

aliafshar/ gemma3-it-qat-tools

12.6K Downloads Updated 9 months ago

Gemma3 instruction-tuned and quantized with tool/functions using a faithful representation of the Google DeepMind recommended tool calling methodology. Runs remarkably well on old GPUs and CPU.

vision tools 1b 4b 12b 27b

ollama run aliafshar/gemma3-it-qat-tools

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

from ollama import chat

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

import ollama from 'ollama'

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

Applications

Claude Code ollama launch claude --model aliafshar/gemma3-it-qat-tools

Codex ollama launch codex --model aliafshar/gemma3-it-qat-tools

OpenCode ollama launch opencode --model aliafshar/gemma3-it-qat-tools

OpenClaw ollama launch openclaw --model aliafshar/gemma3-it-qat-tools

Models

View all →

Name

5 models

Size

Context

Input

gemma3-it-qat-tools:latest

8.9GB · 128K context window · Text, Image · 9 months ago

gemma3-it-qat-tools:latest

8.9GB

128K

Text, Image

gemma3-it-qat-tools:1b

1.0GB · 32K context window · Text · 9 months ago

gemma3-it-qat-tools:1b

1.0GB

32K

Text

gemma3-it-qat-tools:4b

4.0GB · 128K context window · Text, Image · 9 months ago

gemma3-it-qat-tools:4b

4.0GB

128K

Text, Image

gemma3-it-qat-tools:12b

latest

8.9GB · 128K context window · Text, Image · 9 months ago

gemma3-it-qat-tools:12b latest

8.9GB

128K

Text, Image

gemma3-it-qat-tools:27b

18GB · 128K context window · Text, Image · 9 months ago

gemma3-it-qat-tools:27b

18GB

128K

Text, Image

Readme

Gemma3 instruction-tuned and quantized with tool/functions using a faithful representation of the Google DeepMind recommended tool calling methodology. Runs remarkably well on old GPUs and CPU.

References:

See the original Gemma3 information here:

https://ollama.com/library/gemma3