Blog
AI Image & Video API Guides
Hands-on model tests, production tutorials, prompt techniques, and current API pricing for generative media teams.
Recraft V4.1 API Pricing: Flash at $0.01 per Image
recraft-v4.1/text-to-image at $0.05/image for finals and recraft-v4.1-flash/text-to-image at $0.01/image for drafts — identical parameters, model swap only. One same-prompt tier pair (NORTH BEAN poster) plus a similar-prompt product pair, differences stated in the captions. A 9:1 draft-to-final monthly mix saves about 72% versus all-V4.1.
Wan 2.6 Flash Prompts: 4 Templates with Real Clips
Four scenes, all real clips: a fox close-up (default tier), a vertical night market (audio on), a convertible orbit (1080p), and a clear-umbrella push-in (10 s, 3:4). Each scene lists the input image, prompt, and pre-fill link; the aspect follows the image. $0.145 for a default 5-second clip; audio ~2×, 1080p ~1.5×.
How to Use Recraft V4.1 API: On-Brand Images in 5 Minutes
recraft-v4.1/text-to-image ($0.05 finals) and recraft-v4.1-flash/text-to-image ($0.01 drafts) share every parameter — swap the model field only. colors sets the palette and background_color the backdrop (Playground doesn't expose them; API only) as strong guidance. Six size presets; one /v1/tasks flow: submit for a taskId, poll to success, download.
How to Use Krea 3 API: First Image in 5 Minutes
Krea 3 ($0.036/image) reasons about and expands your prompt before generating — short lines still produce full frames; berry/jam variants switch with one field at the same price. The family sibling Krea 2 Medium Turbo ($0.0215/image) offers 8 aspect ratios (incl. 2.35:1 cinema) and 4 creativity levels for volume work. Both run on HiAPI /v1/tasks: submit for a taskId, poll to success, download from output[0].url (temporary — persist it).
How to Use FIBO 1.5 API: Generate and Edit in 5 Minutes
FIBO Gen 1.5 renders photoreal frames with accurate in-image text; FIBO Edit 1.5 changes them with one instruction, up to 4 references, or a mask. The edit field is instruction (not prompt); masks are white-edits, black-keeps. Both run on /v1/tasks at $0.0575: submit for a taskId, poll to success, download.
How to Use Wan 2.6 Flash Image-to-Video API: Aspect Follows Your Image
wan2.6-flash/image-to-video takes one image and a motion line and returns 5/10/15 second 720p/1080p video with optional synchronized audio. Four price combinations: $0.029/s silent 720p, $0.0575 with audio, $0.043 silent 1080p, $0.086 with audio; failures refund. No aspect parameter — the output follows the input image: portrait in, portrait out.
Imagen 4 vs Nano Banana 2.1: Which Should You Use?
Resolution options, references, and jpg / png output point to gemini-nano-banana-2.1 (1K $0.022); fixed tiers with stable materials and lettering point to Imagen 4 (standard $0.06). Two same-prompt comparisons — teapot (1:1) and fisherman (3:4) — with matching aspects and differing formats and pixel sizes. One /v1/tasks endpoint: Imagen 4 requires prompt + aspect_ratio; NB2.1 takes resolution and output_format explicitly.
Imagen 4 API Pricing: $0.03 per Image at Fast
The Imagen 4 family bills per image: imagen-4.0-fast $0.03, imagen-4.0 $0.06, imagen-4.0-ultra $0.09 — price depends on the model tier alone. Aspect ratio, prompt length, and negative_prompt never change the price; failures refund in full. $10 yields about 333 / 166 / 111 images; a 700 Fast + 300 standard mix runs about $39.00 a month.
Use HiAPI with Claude Code: Skip the $100/Month Subscription
Claude Code supports BYOK — you're not locked into the $100/month Max plan Point ANTHROPIC_BASE_URL to HiAPI and use HiAPI's Claude models with pay-as-you-go billing HiAPI's Claude Opus 5.5 costs just $2.02/$10.08 per million tokens — 87% less than typical Opus pricing Step-by-step configuration guide plus model selection advice
One API Key for All AI Models: HiAPI vs the Aggregator Crowd
Several platforms now promise 'one API key for all models' — but most only cover text HiAPI covers 168+ models across text, image, video, audio, and music with a single key Text models range from $0.05/M tokens (GPT-5.6 Luna) to flagship reasoning models Supports Anthropic Messages, OpenAI Chat, and Responses API — minimal migration effort
Build with the models you just read about.
Compare image, video, and audio models, then use the same API workflow in production.