Ollama (Local)

Run translations entirely on your own machine — no internet, no API costs, complete privacy.

Overview

The Ollama backend runs entirely offline on your local machine. It sends the full manga image with a prompt to your local Ollama instance, which returns translated text with position data. This is the most private option — nothing leaves your computer.

Setup

  1. Install Ollama from ollama.ai
  2. Pull a vision-capable model:
    ollama pull gemma3:12b
  3. Start Ollama:
    ollama serve
  4. In the extension popup, go to the Backend tab
  5. Select "Ollama"
  6. Set the URL (default: http://localhost:11434)
  7. Select your model from the dropdown
  8. Click "Check Connection"
Backend tab with Ollama selected and model dropdown

Recommended Models

ModelSizeQualityBest For
gemma3:12b~8GBGoodBalanced quality and speed
llava:7b~4GBDecentLower-end hardware
granite3.2-visionVariesGoodIBM's vision model
qwen3.5:9b~6GBGoodMultilingual tasks
GPU recommended

Vision models are computationally intensive. A GPU with at least 8GB VRAM will provide significantly better translation speed. CPU-only inference works but is much slower.