Install VisionSqueezer (npm, cargo, pip)

VisionSqueezer ships across three registries. Pick whichever fits your workflow — they all expose the same core pipeline.

npm (recommended for MCP)

The npm wrapper downloads the platform-correct prebuilt binary on install. Best path if you want the MCP server in an AI editor.

Prebuilt MCP binaries cover Linux x86_64/ARM64, macOS Intel/Apple Silicon, and Windows x86_64/ARM64. Python wheels cover Linux x86_64/ARM64, macOS Intel/Apple Silicon, and Windows x86_64; other Python architectures can compile from source with pip install --no-binary vision-squeezer vision-squeezer.

npm install -g vision-squeezer

Or run it zero-install via npx. The npm package ships the MCP server plus a one-shot mode that optimizes a file:

npx -y vision-squeezer optimize image.png --max-tokens 1600

That prints a JSON report and writes the optimized copy to a temp file (output_path). For every CLI flag (--model, --format, --dry-run, …) install the vision-squeezer CLI with Cargo below.

Cargo (Rust source)

cargo install vision-squeezer

This builds the vision-squeezer CLI and vision-squeezer-mcp server from source.

Python (pip)

Native wheels via pyo3 + maturin for Linux, macOS, and Windows.

pip install vision-squeezer

import vision_squeezer as vs

report = vs.optimize_image(
    "screenshot.png",
    model="claude",
    auto_quality=0.95,
    output_path="screenshot.optimized.jpg",
)
print(report["tokens_saved"], report["size_reduction_pct"])

See Python Bindings for the full API.

Verify

npx -y vision-squeezer --version          # npm
vision-squeezer --version                 # Cargo CLI
vision-squeezer image.png --model gpt6 --dry-run --json

--dry-run runs the full pipeline without writing to disk or touching the stats database — a safe way to confirm the install works and preview token impact.