Full-Size or Downscaled? A Per-Image Resolution Rule for Opus 4.7's High-Resolution Vision
Opus 4.7 finally read the fine texture in my wallpapers, so I sent everything at full size. My weekly image tokens jumped 2.4x. Here is the preflight that decides resolution and model per image, with the measured savings.
Don't Trust the Confidence Score: Per-Class Calibration and Abstain Routing for Vision Classification
Overall accuracy looked fine while individual categories quietly collapsed. Here is how I calibrated Claude Vision's self-reported confidence per class and routed abstentions to a human queue.
Let Claude Actually See the Images Your Tools Return — Use Image Blocks in tool_result and Cut Tokens by Roughly 10x
Stuffing a base64 string into a tool_result makes the same image cost roughly 10–20x more tokens. Here is how to return it as an image content block instead, with SDK code, a token-cost estimate, and the gotchas I hit in production.
Claude Vision API in Production — Implementation Patterns for Image Analysis, PDF Processing, and OCR
Implementation patterns for taking Claude's vision capabilities to production: choosing between Base64, URL, and the Files API, native PDF processing, schema-enforced extraction with Tool Use, batch cost reduction, and error recovery — all with working code.
Diagnosing invalid_request_error When You Pass an Image URL to the Claude API
When the Claude API rejects an image you passed via `source.type: url`, the root cause almost always lives in one of four buckets: scheme, MIME, size, or reachability. Here is the diagnostic order I use in production.
Automating Wallpaper Classification with Claude Vision API — Accuracy, Cost, and a Batch Job That Survives Interruptions
A firsthand account of automating wallpaper category classification with the Claude Vision API. Honest numbers on accuracy and cost, plus the batch design that stopped me re-paying for the same images.
Pasting Screenshots into Claude Code: A Practical Workflow for Fixing UI Bugs Fast
A field-tested workflow for handing screenshots and design comps to Claude Code from the terminal—covering the three input methods, prompt templates that actually work, and the gotchas that bit me in production.
Build a PDF Analysis and Summarization App with Claude API — Vision plus Extended Thinking
Learn how to build a PDF analysis and summarization application using Claude API's Vision capabilities and Extended Thinking, with step-by-step Python implementation.
Claude API Real-time Multimodal Agent Architecture: Design Patterns & Implementation
Master building real-time multimodal agents combining Vision and Tool Use. Learn streaming pipelines, production error handling, and cost optimization patterns with TypeScript and Python examples.
Multimodal Input Guide — Working with Images and PDFs in the Claude API
How to send images and PDFs to the Claude API, when to reach for the Files API, and the cost and speed lessons you only learn by running it in production — with real measurements.