{"uid":"cap_hBL8AEkL2HtOOeAeMagHQ","slug":"image-analysis-caption-tags-object-detection-00d7e19e","name":"Image Analysis – Caption, Tags & Object Detection","description":"Return a caption, visual tags, and located objects for a supplied image.","url":"https://manyfish-seller.redcliff-d7368db1.westus2.azurecontainerapps.io/api/v1/image/analyze","method":"POST","headers":{},"bodySchema":{"type":"object","properties":{"contentType":{"enum":["image/jpeg","image/png"]},"imageBase64":{"type":"string","pattern":"^(?:[A-Za-z0-9+/]{4})*(?:[A-Za-z0-9+/]{2}==|[A-Za-z0-9+/]{3}=)?$","maxLength":13981016,"minLength":8},"minimumConfidence":{"type":"number","default":0.5,"maximum":1,"minimum":0}}},"responseSchema":{"type":"json","example":{"tags":[{"name":"text","confidence":0.97}],"image":{"width":200,"height":120,"contentType":"image/png"},"caption":{"text":"A red circle and the word hello on a white background.","confidence":0.92},"objects":[],"language":"en"}},"example":null,"exampleRequest":null,"tags":["x402"],"displayCostAmount":"0.004","displayCostAsset":"USDC","priceDynamic":false,"priceHint":null,"priceStatus":"priced","priceSource":"probe","requiresHandshake":false,"reviewCount":0,"rating":{"score":"0.00","successRate":"0.00","reviews":0,"stars":null,"state":"unrated"},"availabilityStatus":"unknown","priceObserved":null,"sessionDeposit":null,"pricing":{"kind":"static","summary":"$0.004/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.004","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.004","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_HKIxvQOmFk9AZA0xdrnn8","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.004","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Analyzes a base64-encoded JPEG or PNG image and returns a descriptive caption, visual tags, and detected objects with confidence scores.","exampleAgentPrompt":"Analyze this JPEG image for me — give me a caption describing the scene, any relevant visual tags, and a list of objects you can detect, only including results with at least 70% confidence.","exampleUseCases":[{"title":"E-commerce product image tagging","prompt":"I have a PNG photo of a running shoe I want to list online — can you analyze it and give me a caption, visual tags, and any detected objects so I can auto-populate the product description?"},{"title":"Content moderation pre-screening","prompt":"Before I store this user-uploaded JPEG, analyze it and tell me what objects and tags are detected — only show me things with at least 80% confidence so I know what's in the image."},{"title":"Accessibility alt-text generation","prompt":"Here's a base64-encoded JPEG of our blog hero image — analyze it and return a caption and visual tags so I can write an accurate alt-text for screen readers."}],"resultDescription":"Returns a natural-language caption describing the image, a list of visual tags (keywords describing scene elements, colors, styles, etc.), and a list of detected objects each with a label and confidence score. Results are filtered by the supplied minimum confidence threshold (default 0.5).","failureModes":["Invalid base64 string — malformed or non-base64 payload returns a 400 error","Unsupported content type — only image/jpeg and image/png are accepted","Image too large — base64 payload exceeding ~10MB may be rejected","Low-quality or ambiguous images may return sparse tags or a generic caption","Confidence threshold set too high may return empty object lists"],"whenToPreferThis":"Choose this endpoint when you need a combined response of caption, tags, AND object locations in a single call, rather than separate endpoints for each task. It is particularly useful for enriching image metadata, powering search indexing, or feeding downstream NLP pipelines. Prefer the dedicated object-detection endpoint if you only need bounding boxes, or the caption-only endpoint if you need a single sentence description.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-09-14T12:40:29.166Z","isFirstParty":false}