Model APIs

GPT Image 2.5 Sunburst Text to Image, OpenAI's precision-focused GPT Image 2.5 tier for everyday text-to-image generation. Five quality tiers (low to max), strong prompt fidelity and accurate in-image text. Billed per output image by quality and image size.

GPT Image 2.5 Sunburst Image to Image, OpenAI's precision-focused GPT Image 2.5 tier for natural-language image editing. Accepts up to 16 reference images, supports mask for scoped edits, transparent background, five quality tiers. Billed per output image by quality and image size, plus per additional input image.

GPT Image 2.5 Sunburst Text to Image, OpenAI's GPT Image 2.5 Sunburst tier for text-to-image generation, served via a low-price with flat per-call pricing. Exposes resolution and aspect ratios. The edition is priced lower than the official version.

GPT Image 2.5 Sunburst Image to Image, OpenAI's GPT Image 2.5 Sunburst tier for natural-language image editing, served via a low-price with flat per-call pricing. Accepts up to 10 reference images. The edition is priced lower than the official version.

GPT Image 2.5 Flare Text to Image, OpenAI's fast, balanced GPT Image 2.5 tier for everyday text-to-image generation. Five quality tiers (low to max), strong prompt fidelity and accurate in-image text. Billed per output image by quality and image size.

GPT Image 2.5 Flare Image to Image, OpenAI's fast, balanced GPT Image 2.5 tier for natural-language image editing. Accepts up to 16 reference images, supports mask for scoped edits, transparent background, five quality tiers. Billed per output image by quality and image size, plus per additional input image.

GPT Image 2.5 Flare Text to Image, OpenAI's GPT Image 2.5 Flare tier for text-to-image generation, served via a low-price with flat per-call pricing. Exposes resolution and aspect ratios. The edition is priced lower than the official version.

GPT Image 2.5 Flare Image to Image, OpenAI's GPT Image 2.5 Flare tier for natural-language image editing, served via a low-price with flat per-call pricing. Accepts up to 10 reference images. The edition is priced lower than the official version.

Alibaba Happy Horse 1.1 Image to Video is Alibaba's Image to video model (v1.1), generating cinematic video based on the first frame at 480p/720p/1080p with smooth camera movement and expressive motion. Billed per second of output video by resolution tier.

Alibaba Wan 3.0 FLF to Video is Alibaba's FLF to video model generating high-quality video from first and last frames at 480p/720p/1080p with flexible duration and aspect ratio options. Billed per second of output video by resolution tier.

ByteDance Seedream 5.0 Pro Image to Image Layer Decomposition, ByteDance Seedream 5.0 Pro Layer Decomposition is ByteDance's image layer separation model that decomposes a generated image into multiple layers for professional design and compositing workflows. Billed per generated layer by pixel tier.

ByteDance Seedream 5.0 Pro Image to Image Edit is ByteDance's premium image editing model that transforms existing images via natural language instructions, supporting multi-reference image input with consistent subject identity. The first input image is free; subsequent images are $0.003 each. Output is billed per image by pixel tier.

ByteDance Seedance 2.5 Video Extend is ByteDance's video extend model that extends an existing video with a prompt-driven cinematic continuation generated seamlessly from its last frame. Billing is based on the combined duration of the reference video and the new segment by resolution tier; the reference duration is clamped to 2-30 seconds and the new segment duration is set by the duration parameter.

ByteDance Seedance 2.5 Video Edit is ByteDance's video editing model that applies prompt-driven edits to an existing video while maintaining motion and temporal consistency. Billing is based on the combined duration of input and output video by resolution tier; the input video is handled within the 4-30s range before pricing.

ByteDance Seedance 2.5 Text to Video is ByteDance's latest video generation model supporting up to 4K resolution with native audio, cinematic camera control, and multi-character consistency. Billed per second of output video by resolution tier.

ByteDance Seedance 2.5 Reference to Video is ByteDance's reference-to-video model supporting multimodal references. When reference videos are provided, billing duration includes both input and output video duration at a discounted rate. A minimum total duration rule applies to short reference clips.