Blog
AI Image & Video API Guides
Hands-on model tests, production tutorials, prompt techniques, and current API pricing for generative media teams.
Seedream 5.0 Lite API Guide: Text-to-Image and Image Editing
Two endpoints, one price: $0.035 per image at 2K or 4K. Strong at exact text in images and targeted edits. Image-to-image takes up to 14 references at no extra cost.
How to Use the Nano Banana 2 Lite API: curl, Python, and a Working Request
Nano-Banana-2-Lite runs on hiapi's unified async task API: POST /v1/tasks, poll GET /v1/tasks/{id}, download data.output[0].url before it expires. Required input is prompt + aspect_ratio; the model id is case-sensitive (lowercase returns 400 MODEL_UNAVAILABLE). Reference images go in image_urls (max 10) — unlike Nano-Banana-2/Pro which use image_input; this rename is the main porting gotcha. Lite accepts no resolution/seed/n fields; step up to Nano-Banana-2 or Pro for 1K/2K/4K control. For production use callback.url with when:'final' instead of polling, persist taskId for idempotent retries, and store image bytes rather than expiring URLs.
Qwen Image 2.0 and 2.0 Pro API: Prices, Real Outputs, Code
Two tiers with the same request: qwen-image-2.0 at $0.0422 and qwen-image-2.0-pro at $0.107. Every requested Chinese and English string in our tests rendered correctly. Qwen Image 3.0 ($0.0358) and 2.1 Turbo ($0.02) now cost less than 2.0 and can edit images.
Short-Form Video with seedance-2.0-mini: An End-to-End hiapi API Workflow
End-to-end workflow for TikTok/Reels-style vertical clips with seedance-2.0-mini on hiapi: first curl call, reusable Python client, I2V first-frame control, and a batch script for a week of content. Verified against the live API: prompt, duration (4-15s) and resolution (480p/720p) are required; aspect_ratio supports 9:16 for vertical; the schema is strict and rejects unknown fields. 480p drafts cost $0.0272/second (~$0.11 for 4s), so the draft-at-480p / re-render-keepers-at-720p loop stays cheap; output MP4s include native synced AAC audio. The demo clips embedded in the article are real outputs from the exact requests shown, including a vertical 9:16 clip with native audio.
Best Image-to-Image API in 2026: Same Edit, Three Models, Real Results
We ran the identical background-swap edit through gpt-image-2/image-to-image@ext, nano-banana-2, and nano-banana-pro using one source image and one verbatim prompt Nano Banana 2 and Pro preserved the product's exact matte glaze and shape; gpt-image-2 followed the instruction fully but re-rendered the glaze glossier and added a tan rim - identity drift gpt-image-2 i2i starts at $0.007/image (low, 1K) and tops at $0.76 (high, 4K); nano-banana-2 runs $0.076-0.114; nano-banana-pro $0.17-0.2992 Input schemas differ per family: image_urls + required quality/resolution for GPT Image 2, image_input + aspect_ratio for Nano Banana All three run through the same POST /v1/tasks async endpoint, so re-running the test on your own image is three POSTs with the model string swapped
Wan 2.7 Text-to-Image for E-Commerce Product Photos: 4K Studio Shots via One API Call
wan2.7-image/text-to-image generates up to 4096x4096 product photos at a flat $0.08 per image on hiapi, with no 4K surcharge. The async task API returns in 60-90 seconds per image: POST /v1/tasks, poll the task ID, download output[0].url before it expires. Aspect ratios are strictly limited to 1:1, 16:9, 4:3, 21:9, 3:4, 9:16, 8:1, 1:8 — common photo ratios like 3:2 and 4:5 return 400 INVALID_REQUEST. Short quoted brand text renders reliably on packaging shots, but unspecified labels come out as gibberish — quote exact strings or prompt for unlabeled products.
Using Nano-Banana for E-Commerce Product Images via the hiapi API
An anchor-and-restage e-commerce workflow on hiapi: nano-banana generates the $0.021 white-background listing shot, then Nano-Banana-2 restages that exact bottle — and even the same human model — into lifestyle, packaging, and campaign scenes via the image_input reference array. Every image was generated during writing with the exact prompt shown, and all prices are checked against the live pricing page.
Grok Imagine Video API: 1.0 and 1.5 Image-to-Video, Text-to-Video
Three endpoints at the same price: $0.015/s at 480p, $0.0282/s at 720p. 1.0 runs 6–30 s with up to 7 references; 1.5 runs 1–15 s from one image. Every clip comes with generated sound; there's no switch.
Short-Form Video with wan2.7-video/text-to-video@pro: An End-to-End hiapi API Workflow
wan2.7-video/text-to-video@pro generates up to 1080P MP4 clips with native AAC audio, 2-15 seconds, in five aspect ratios including vertical 9:16 — all through hiapi's async /v1/tasks endpoint. The guide includes a verified full input schema, working curl and Python examples, and two real clips generated by the exact code shown.
How to Use seedance-2.0-fast via the hiapi API: curl, Python, and a Working Request
seedance-2.0-fast is a text-to-video model on hiapi's unified async /v1/tasks endpoint: POST to create a task, poll GET /v1/tasks/<id> or register a callback, then download data.output[0].url before the signed link expires. The input schema is strict — only prompt (required), duration (integer 4–15), resolution (480p/720p), and aspect_ratio are accepted; every request shape and error message in this guide was verified against the live API.
Build with the models you just read about.
Compare image, video, and audio models, then use the same API workflow in production.