Gemini 3.8 Flash (vision)
Understand images, videos and public YouTube links. Extract text, compare images, summarize clips and return text or JSON with usage-based pricing.
Gemini 3.8 Flash understands existing images, video and public YouTube videos. Send media and an explicit prompt to describe scenes or products, read visible text (OCR), compare up to 8 images, extract structured fields, summarize or transcribe a video, or identify key moments with timestamps. Give exactly one media source: media_url (a public HTTPS image or video URL, a YouTube link, or a JSON list of image URLs), an image upload or a video upload; an upload wins over a link. Set kind='video' for an extension-less video link. Images: 20 MiB each; video: 512 MiB, one per call. Input is limited to 200,000 tokens, roughly 13 minutes of video. max_output_tokens is 256-65,536 (default 8,192). Controls: response_schema (a JSON Schema the answer must follow; implies JSON), thinking_level (low, medium or high; low makes labeling cheaper and faster), media_resolution (low, medium or high; low spends fewer input tokens on an image, high costs a video about 3x), temperature and seed, and for a video fps, start_offset and end_offset in seconds. Returns analysis as a string (parse it for JSON), model, kind, source, usage (input, output and cached tokens; thinking_tokens are part of output_tokens), finish_reason (MAX_TOKENS means the answer was cut), request_id and charge_usd. The Models API also returns calculated_price (equal to charge_usd) and, when a field is not a control of this model, warnings naming each ignored field. Billing uses actual input and output tokens including thinking. It does not create or edit media or read private or login-only YouTube videos. Recognition and transcription may be imperfect; review critical extracted facts.
- Visual Language Models (VLMs)
Price: $0.75 per 1M input tokens, $3.75 per 1M output tokens, cached input $0.075 per 1M.
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