All AI models on FlyAIgh

OpenAI's flagship text-to-video model. Cinematic realism, native audio, 4–12s clips up to 1080p.

Google DeepMind's VEO 3.1 — three tiers of 8-second video with native audio, image-to-video and first/last frame control.

ByteDance's flagship video model — cinematic camera work, native audio, 4–15 second clips with image-to-video and multi-image reference.

Kuaishou's Kling V3 — strong on character motion and reference-driven scenes. Multi-image reference up to 7 inputs on Kling V3 Omni.

MiniMax Hailuo 2.3 — SOTA instruction following with 15 named camera commands. 6 or 10 second clips at 768p or 1080p.

xAI's video model — the cheapest flagship on FlyAIgh at 2 credits per second. Natural motion dynamics, native audio, 6–15 second clips.

Alibaba Bailian's HappyHorse — auto-routing video model with reference-to-video and up to 9 input images. The latest 1.1 sharpens visuals, smooths motion, and holds character consistency across shots. 3–15 second clips at 1080p.

Alibaba's WAN 2.7 video model — multi-shot narrative generation with full mode coverage (t2v / i2v / r2v / edit). Up to 1080p, 2–15s clips.

ShengShu Technology's Vidu Q3 Pro — full-mode video coverage with up to 7 reference images. Fixed 4 / 5 / 8 second clips at 720p.

Pixverse — v6 for general-purpose video and C1 for action / effects / fast-motion. Fixed 5 / 8 second clips at 720p with multi-shot control.

Google's image generation family — Nano Banana 2 for fast everyday image work and Nano Banana Pro for 4K output and AI image editing.

OpenAI's image generation model — best-in-class text rendering inside images and strong image editing support.

DeepSeek's professional image model — 4K output with Thinking Mode for prompt analysis, watermark control and up to 9 reference images.

Midjourney's iconic aesthetic — the new v8.1 sharpens detail and prompt adherence, alongside v7. Three speed tiers (Draft / Fast / Turbo), seven aspect ratios, and image-to-image support.

ByteDance's Seedream 5.0 Lite — versatile photorealistic image model with seven aspect ratios including ultrawide 21:9 / 9:21.

Z.ai's Z Image Turbo — the cheapest image model on FlyAIgh at just 1 credit per image. Fast turnaround, 1K or 1.5K output.

Alibaba's Qwen Image 2.0 — strongest Chinese + English text rendering on FlyAIgh, unified text-to-image and image-editing in one model. 2.0 and Pro variants.

Kuaishou's Kling Image 3.0 — strongest character consistency on FlyAIgh. 3.0 for text-to-image, 3.0 Omni for multi-image reference with up to 7 inputs.