Every model, one subscription
55 generation models in one place — 27 video, 23 image and 5 audio. Your credits work across all of them, and higher tiers get priority queues.
Not sure which to pick? Start with Seedance 2.0 or Kling 3.0 for video, GPT Image 2 or Nano Banana Pro for images — then compare results against the rest. Every generation is billed in credits by model and output, so trying a different model is just another run, not another subscription.
Video models (27)
Generate video →Seedance 2.5
720p4–30sNext-gen text/image/reference/audio to video, up to 30s and 30 reference images
Seedance 2.5 II
720p5–10sNext-gen text/image/reference/audio to video, smoother motion & camera control
MiniMax H3
2K4–15sNext-gen Hailuo text/image to video, 4–15s, up to 2K
Seedance 2.0
4K4–15sText/image/reference/audio to video, adaptive ratio
Seedance 2.0 Fast
720p4–15sTurbo Seedance 2.0
Seedance 2.0 Mini
720p4–15sLightweight Seedance 2.0
Seedance 1.5 Pro
1080p4–12sFirst/last-frame conditioned generation
Happy Horse 1.1
1080p3–15sText/image/reference to video
Wan 2.7 II
1080p2–15sImage to video (first frame, optional audio drive), 2–15s, ratio follows source
Kling 3.0
5–10sText/image to video, up to 7 references
Kling 2.6
5–10sText/image to video
Kling V3 Turbo
1080p5–10sText/image to video, 720p/1080p
Kling 2.5 Turbo Pro
5–10sText/image to video, first/last frame
Kling 2.1
5–10sImage to video
Seedance 2.0 II
4K5–10sText/image/reference/audio to video
Seedance 2.0 Fast II
720p5–10sTurbo edition
Seedance 2.0 Mini II
720p4–15sLightweight, 4–15s
Veo 3.1
1080p5–8sImage to video, up to 7 images
Veo 3 Fast
5–8sText to video
Wan 2.7
1080p5–10sText/image to video, optional audio
Gemini Omni Video
4K4–10sText/image to video, up to 4K
Grok Video
720p6–10sText/image to video
Imagen Video 1.5
1080p5–8sText/image to video
Hailuo 2.3
5–10sImage to video
Hailuo 2.3 Fast
5–10sImage to video (turbo)
Happy Horse 1.0
1080p5–10sText/image to video
OmniHuman 1.5 (数字人)
1080Audio-driven talking avatar
Image models (23)
Generate images →Qwen Image 3.0
2KText/image to image, accurate CN/EN text rendering, same price at 1K and 2K
Qwen Image 3.0 Pro
2KEnhanced Qwen Image 3.0, steadier detail and text
Seedream 5.0
4KHigh-quality text to image
Z-Image Spicy
1KText to image without content pre-filter — produces the first frame for editing and image-to-video
Nano Banana Pro
4KText/image generation, 1K–4K
Nano Banana 2
4KText/image generation, 1K–4K
Nano Banana
Text/image generation
Seedream 4.5
4KText/image generation, basic/high
Seedream 5.0 Lite
4KText/image generation, 2K/4K
Seedream 4.0
4KText/image generation, 1K–4K
ByteDance Seedream
Text to image
Flux 2 Pro
2KText to image, 1K–2K
Flux 2 Flex
2KText to image, 1K–2K
GPT Image 2
4KText to image, 1K–4K
GPT Image 1.5
Text to image, medium/high
Imagen 4
Text to image
Ideogram V3
Text to image, negative prompt support
Recraft V3
Text to image
Z-image Turbo
Text to image (turbo)
Runway Gen-4
Text/image generation
Grok Imagine
Image editing (source image required)
Qwen Image Edit
Image editing (source image required)
Qwen Image Edit II
Image editing (one-line instruction, async, ratio follows source)
Audio models (5)Music & voiceover
Create audio →Suno V5
MusicText to music — full songs or instrumentals, vocals included
Suno V4.5
MusicText to music (previous gen, great value)
Eleven Multilingual v2
VoiceMultilingual voiceover & narration, lifelike delivery
Eleven Turbo v2.5
VoiceFast, low-latency text to speech
Eleven Dialogue v3
VoiceMulti-speaker dialogue speech