Models

Multi-modal video generation — text, image, video, audio inputs

Faster Seedance 2.0 with lower pricing — same capabilities, quicker generation

Google Veo 3.1 — text-to-video, image-to-video, first & last frame, with background music.

OpenAI Sora 2 — high-quality text-to-video and image-to-video generation.

Kuaishou Kling 3.0 — text-to-video and image-to-video with native audio, 720P/1080P, 3-15 sec.

Kuaishou Kling 2.6 — image-to-video, 5s/10s, with optional native audio.

Google Gemini 3.1 Flash Image — lightning-fast speed with Pro-level quality, 4K output, and text rendering.

OpenAI GPT Image 2 — one public model id for text-to-image and image-to-image editing.

Ideogram V4 — text-to-image with crisp visuals and accurate text rendering, ideal for logos, posters, and brand assets.

OpenAI GPT 5.4 — multimodal reasoning model, 256K context, Responses API.

Anthropic Claude Opus 4.6 — flagship reasoning model, 200K context, optional extended thinking.

Anthropic-compatible Claude family — Haiku, Sonnet, and Opus models with 1M context and optional extended thinking.