GPT Image 2
OnlineText-to-image with accurate text rendering and layout-heavy compositions.
gpt-image-2 Models Overview
All models — image, video, and audio — are called through the unified POST /v1/tasks endpoint, with the model selected via the model field and parameters in input. See each model page for its input fields.
Current pricing is maintained on the main site: HiAPI Pricing.
Image Models
Text-to-image, image-to-image, reference editing, and photorealistic generation.
GPT Image 2 Pro
OnlineHigher-fidelity text-to-image for brand visuals, English prompts, and 2K output.
gpt-image-2-pro GPT Image 2 Image-to-Image
OnlineImage-to-image generation from one or more reference images.
gpt-image-2-image-to-image GPT Image 2 Pro Image-to-Image
OnlinePro reference-image editing and redraws from 1-5 images, up to 2K output.
gpt-image-2-image-to-image-pro GPT Image 2 Beta
BetaPreview image generation model for early testing.
gpt-image-2-beta Nano Banana
OnlineFast, prompt-tolerant text-to-image for quick iteration.
Nano-Banana Nano Banana 2
OnlineBalanced Nano Banana tier with optional references and 1K / 2K / 4K output.
Nano-Banana-2 Nano Banana Pro
OnlinePremium brand visuals, reference-based editing, and high-resolution output.
Nano-Banana-Pro FLUX 1.1 Pro
OnlinePhotorealistic image generation with optional reference-image guidance.
flux-1.1-pro Qwen Image 2.0
OnlineLow-cost image model with strong Chinese prompt and text rendering behavior.
qwen-image-2.0 Video Models
Text-to-video, image-to-video, reference media, and short-form generation.
Seedance 2.0
OnlineReference images, video, and audio, with optional synchronized audio.
seedance-2-0 Wan 2.7 Text-to-Video
OnlineText-to-video with size, duration, prompt expansion, and shot controls.
wan2.7-t2v Wan 2.7 Image-to-Video
OnlineAnimate a first frame, with optional last frame, audio, or clip inputs.
wan2.7-i2v HappyHorse 1.0
OnlineShort text-to-video clips, 3-15 seconds, at 720p or 1080p.
happyhorse-1-0 Audio
Audio generation models will be documented once available.
Audio models
Coming soonNo public audio model ID is available yet. The docs page will be updated after launch.
No model ID yet