$ loading_
Snap images to model tile boundaries to reduce vision LLM token costs.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "io.github.eralpozcan/vision-squeezer" MCP server from askskill: Run: claude mcp add 'io-github-eralpozcan-vision-squeezer' -- npx -y vision-squeezer
Please snap this batch of screenshots for Claude and GPT-4o to vision tile boundaries, minimize image token usage without hurting readability, and return the optimized file list with estimated savings.
Returns processed images, resized dimensions, and estimated token or cost savings for each image.
Help me integrate vision-squeezer into the current image upload flow: optimize images using Gemini and Qwen tile rules before sending them to downstream recognition services, and explain the integration steps.
Provides an integration-ready processing flow, configuration guidance, and before/after cost impact details.
Batch process this dataset of images into more token-efficient sizes for Llama and Qwen vision models, preserve key content, and generate a processing report.
Outputs optimized dataset images and a report covering dimension changes, target models, and estimated savings.
Compress LLM text and JSON to reduce tokens and lower usage costs.
Analyze images, extract text, and answer visual questions with LLM vision models.
Connect local vision models for image analysis, comparison, and OCR.
Run local OCR and image analysis on macOS using Apple Vision.
Connect vision models via MCP for image description and compression tasks.
Access institutional AI market intelligence, council verdicts, squeeze scans, and options flow.