Gemini 3.8 Flash (vision)
Gemini 3.8 Flash reads an image or a video and answers in text or JSON. Give it one or more image URLs, a direct video URL or a YouTube link, plus a question: what is shown, what is said, what text appears on screen, which products or people are present, the style and the key moments with timestamps. Use it to describe a product photo, pull structured fields out of a screenshot or receipt, check a creative against a brief, or summarise a clip. Media arrives as a public HTTPS URL - an AgentFile public_url, a CDN link or a YouTube link; nothing is stored beyond the request. Answers are text unless response_format is json, and billing follows the tokens Gemini actually read and wrote, so a long video costs more than a photo. It does not generate or edit images and video (use the media models for that), and it cannot open pages behind a login.
Price: $0.8625 per million input tokens; $4.3125 per million output tokens. Charged from actual usage.
Run Gemini 3.8 Flash (vision) via API
curl -X POST https://api.flymy.ai/api/v1/flymyai/gemini-3_8-flash-vlm/predict \
-H "X-API-KEY: $FLYMYAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{}'API documentation · Create a free account · Pricing