# Neon AI model and workflow catalog ## Trending fixed-template workflow Kie Seedance 2.5 credential readiness: configured. Configured credentials are not a provider health guarantee. Available profiles use 480p, one pinned source and one image per selected role. Source length is retained, with adaptive output duration; actual results can differ. Soundtrack is restored where the selected source supplies one; Rumpelstiltskin portrait is silent. There are 23 themes and 32 comparison profiles; the two source-only themes cannot be submitted. Canonical: https://neonai.io/models.md. Pricing: https://neonai.io/pricing.md. Video workspace: https://neonai.io/ai-video-generator. Image workspace: https://neonai.io/ai-image-generator. Only application integrations are listed here, not every SEO/alternative page as an independent model. Enabled means its required provider credential is configured. Provider availability, policy and final server input validation can still reject requests. Model names are integration labels; this document does not assert provider affiliation, benchmark quality or native output pixels beyond the exposed workspace parameters. ## Currently enabled video integrations ### Face Swap Motion Transfer (genjutsu-motion-transfer) Provider label: Higgsfield. Modes: reference. Operation: motion-transfer. Prompt: optional; maximum 10000 characters. Input maxima by mode: {"reference":{"images":8,"videos":1,"audio":0,"total":12}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480p, 720p, 1080p. Durations: {"min":4,"max":30,"step":1}. Aspect ratios by mode: {"reference":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"720p","duration":4,"aspectRatio":"adaptive","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Face Swap Object Swap (genjutsu-object-swap) Provider label: Higgsfield. Modes: reference. Operation: object-swap. Prompt: optional; maximum 10000 characters. Input maxima by mode: {"reference":{"images":8,"videos":1,"audio":0,"total":12}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480p, 720p, 1080p. Durations: {"min":4,"max":30,"step":1}. Aspect ratios by mode: {"reference":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"720p","duration":4,"aspectRatio":"adaptive","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Face Swap Restyle (genjutsu-restyle) Provider label: Higgsfield. Modes: reference. Operation: restyle. Prompt: optional; maximum 10000 characters. Input maxima by mode: {"reference":{"images":5,"videos":1,"audio":0,"total":12}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480p, 720p, 1080p. Durations: {"min":4,"max":30,"step":1}. Aspect ratios by mode: {"reference":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"720p","duration":4,"aspectRatio":"adaptive","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### MiniMax H3 Max Turbo (minimax-h3-max-turbo) Provider label: MiniMax. Modes: text, image. Operation: generation. Prompt: required; maximum 50000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480P, 768P, 1080P. Durations: {"min":5,"max":15,"step":1}. Aspect ratios by mode: {"text":["21:9","16:9","4:3","1:1","3:4","9:16"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"480P","duration":5,"aspectRatio":"16:9","audio":true}. Audio: native. End image: true. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### MiniMax H3 (minimax-h3) Provider label: MiniMax. Modes: reference, text, image. Operation: generation. Prompt: required; maximum 7000 characters. Input maxima by mode: {"reference":{"images":9,"videos":3,"audio":3,"total":12},"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 2K. Durations: {"min":5,"max":15,"step":1}. Aspect ratios by mode: {"reference":["adaptive","21:9","16:9","4:3","1:1","3:4","9:16"],"text":["21:9","16:9","4:3","1:1","3:4","9:16"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"2K","duration":5,"aspectRatio":"adaptive","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: true. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### H3 Max (minimax-h3-max) Provider label: MiniMax. Modes: text, image. Operation: generation. Prompt: required; maximum 20000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480P, 768P. Durations: {"min":5,"max":15,"step":1}. Aspect ratios by mode: {"text":["21:9","16:9","4:3","1:1","3:4","9:16"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"768P","duration":5,"aspectRatio":"16:9","audio":true}. Audio: native. End image: true. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Seedance 2.5 (seedance-2.5) Provider label: Seedance. Modes: reference, text, image. Operation: generation. Prompt: required; maximum 20000 characters. Input maxima by mode: {"reference":{"images":30,"videos":10,"audio":10,"total":50},"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480p, 720p. Durations: {"min":4,"max":30,"step":1}. Aspect ratios by mode: {"reference":["adaptive","16:9","9:16","1:1","4:3","3:4","21:9"],"text":["adaptive","16:9","9:16","1:1","4:3","3:4","21:9"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"480p","duration":4,"aspectRatio":"adaptive","audio":true}. Audio: optional. End image: true. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Seedance 2.0 Fast (seedance-2.0-fast) Provider label: Seedance. Modes: reference, text, image. Operation: generation. Prompt: required; maximum 20000 characters. Input maxima by mode: {"reference":{"images":9,"videos":3,"audio":3,"total":12},"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480p, 720p. Durations: {"min":4,"max":15,"step":1}. Aspect ratios by mode: {"reference":["adaptive","16:9","9:16","1:1","4:3","3:4","21:9"],"text":["adaptive","16:9","9:16","1:1","4:3","3:4","21:9"],"image":["adaptive","16:9","9:16","1:1","4:3","3:4","21:9"]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"480p","duration":4,"aspectRatio":"adaptive","audio":true}. Audio: optional. End image: true. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Seedance 2.0 Pro (seedance-2.0-pro) Provider label: Seedance. Modes: reference, text, image. Operation: generation. Prompt: required; maximum 20000 characters. Input maxima by mode: {"reference":{"images":9,"videos":3,"audio":3,"total":12},"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480p, 720p, 1080p, 4k. Durations: {"min":4,"max":15,"step":1}. Aspect ratios by mode: {"reference":["adaptive","16:9","9:16","1:1","4:3","3:4","21:9"],"text":["adaptive","16:9","9:16","1:1","4:3","3:4","21:9"],"image":["adaptive","16:9","9:16","1:1","4:3","3:4","21:9"]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"480p","duration":4,"aspectRatio":"adaptive","audio":true}. Audio: optional. End image: true. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Seedance 2.0 Mini (seedance-2.0-mini) Provider label: Seedance. Modes: reference, text, image. Operation: generation. Prompt: required; maximum 20000 characters. Input maxima by mode: {"reference":{"images":9,"videos":3,"audio":3,"total":12},"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480p, 720p. Durations: {"min":4,"max":15,"step":1}. Aspect ratios by mode: {"reference":["adaptive","16:9","9:16","1:1","4:3","3:4","21:9"],"text":["adaptive","16:9","9:16","1:1","4:3","3:4","21:9"],"image":["adaptive","16:9","9:16","1:1","4:3","3:4","21:9"]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"480p","duration":4,"aspectRatio":"adaptive","audio":true}. Audio: optional. End image: true. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Seedance 1.5 Pro (seedance-1.5-pro) Provider label: Seedance. Modes: text, image. Operation: generation. Prompt: required; maximum 20000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480p, 720p, 1080p. Durations: [4,8,12]. Aspect ratios by mode: {"text":["16:9","9:16","1:1","4:3","3:4","21:9"],"image":["16:9","9:16","1:1","4:3","3:4","21:9"]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"480p","duration":4,"aspectRatio":"16:9","audio":false}. Audio: optional. End image: true. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Wan 2.7 (wan-2.7) Provider label: Wan. Modes: text, image. Operation: generation. Prompt: required; maximum 5000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 720p, 1080p. Durations: {"min":2,"max":15,"step":1}. Aspect ratios by mode: {"text":["16:9","9:16","1:1","4:3","3:4"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"720p","duration":5,"aspectRatio":"16:9","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: true. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Wan 2.6 (wan-2.6) Provider label: Wan. Modes: text, image. Operation: generation. Prompt: required; maximum 1500 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":1,"videos":0,"audio":0,"total":1}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 720p, 1080p. Durations: [5,10,15]. Aspect ratios by mode: {"text":[],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"720p","duration":5,"aspectRatio":"16:9","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Grok Imagine (grok-imagine) Provider label: Grok. Modes: text, image. Operation: generation. Prompt: required; maximum 5000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":1,"videos":0,"audio":0,"total":1}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 720p. Durations: [6,10]. Aspect ratios by mode: {"text":["16:9","9:16","1:1","3:2","2:3"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"720p","duration":6,"aspectRatio":"16:9","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: ["normal","fun","spicy"]. Output formats: ["mp4"]. ### Kling 3.0 (kling-3.0) Provider label: Kling. Modes: text, image. Operation: generation. Prompt: required; maximum 2500 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 720p. Durations: {"min":3,"max":15,"step":1}. Aspect ratios by mode: {"text":["16:9","9:16","1:1"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"720p","duration":3,"aspectRatio":"16:9","audio":false}. Audio: optional. End image: true. Seed: false. Web search: false. Last frame: false. Versions: ["standard","pro"]. Styles: []. Output formats: ["mp4"]. ### Seedance 2.5 Face Swap (kie-seedance-2-5-face-swap) Provider label: ByteDance. Modes: reference. Operation: swap. Prompt: not used; maximum 2000 characters. Input maxima by mode: {"reference":{"images":1,"videos":1,"audio":0,"total":12}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480p, 720p, 1080p. Durations: [5]. Aspect ratios by mode: {"reference":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"720p","duration":5,"aspectRatio":"adaptive","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### PixVerse Extend (pixverse-extend) Provider label: PixVerse. Modes: reference. Operation: extend. Prompt: required; maximum 2000 characters. Input maxima by mode: {"reference":{"images":0,"videos":1,"audio":0,"total":12}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 360p, 540p, 720p, 1080p. Durations: [5]. Aspect ratios by mode: {"reference":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"720p","duration":5,"aspectRatio":"adaptive","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### ByteDance Video Upscaler (bytedance-video-upscaler) Provider label: ByteDance. Modes: reference. Operation: upscale. Prompt: not used; maximum 2000 characters. Input maxima by mode: {"reference":{"images":0,"videos":1,"audio":0,"total":12}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 1080p, 2k, 4k. Durations: [5]. Aspect ratios by mode: {"reference":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"1080p","duration":5,"aspectRatio":"adaptive","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Bria Video Background Removal (bria-video-background) Provider label: Bria. Modes: reference. Operation: remove-background. Prompt: not used; maximum 2000 characters. Input maxima by mode: {"reference":{"images":0,"videos":1,"audio":0,"total":12}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: source. Durations: [5]. Aspect ratios by mode: {"reference":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"source","duration":5,"aspectRatio":"adaptive","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### PixVerse Person Swap (pixverse-swap) Provider label: PixVerse. Modes: reference. Operation: swap. Prompt: not used; maximum 2000 characters. Input maxima by mode: {"reference":{"images":1,"videos":1,"audio":0,"total":12}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 360p, 540p, 720p. Durations: [5]. Aspect ratios by mode: {"reference":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"720p","duration":5,"aspectRatio":"adaptive","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Sync Lip Sync 2 (sync-lipsync-2) Provider label: Sync. Modes: reference. Operation: lipsync. Prompt: not used; maximum 2000 characters. Input maxima by mode: {"reference":{"images":0,"videos":1,"audio":1,"total":12}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: source. Durations: [5]. Aspect ratios by mode: {"reference":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"source","duration":5,"aspectRatio":"adaptive","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Sync Talking Photo (sync-talking-photo) Provider label: Sync. Modes: reference. Operation: talking-photo. Prompt: not used; maximum 2000 characters. Input maxima by mode: {"reference":{"images":1,"videos":0,"audio":1,"total":12}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: source. Durations: [5]. Aspect ratios by mode: {"reference":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"source","duration":5,"aspectRatio":"adaptive","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Hailuo 02 Pro (hailuo-02-pro) Provider label: MiniMax. Modes: text, image. Operation: generation. Prompt: required; maximum 2000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 1080p. Durations: [6]. Aspect ratios by mode: {"text":[],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"1080p","duration":6,"aspectRatio":"16:9","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: true. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Hailuo 02 Standard (hailuo-02-standard) Provider label: MiniMax. Modes: text, image. Operation: generation. Prompt: required; maximum 2000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 768p. Durations: [6,10]. Aspect ratios by mode: {"text":[],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"768p","duration":6,"aspectRatio":"16:9","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: true. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Hailuo 2.3 Pro (hailuo-2.3-pro) Provider label: MiniMax. Modes: text, image. Operation: generation. Prompt: required; maximum 2000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":1,"videos":0,"audio":0,"total":1}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 1080p. Durations: [6]. Aspect ratios by mode: {"text":[],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"1080p","duration":6,"aspectRatio":"16:9","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Hailuo 2.3 Standard (hailuo-2.3-standard) Provider label: MiniMax. Modes: text, image. Operation: generation. Prompt: required; maximum 2000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":1,"videos":0,"audio":0,"total":1}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 768p. Durations: [6,10]. Aspect ratios by mode: {"text":[],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"768p","duration":6,"aspectRatio":"16:9","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Seedance 1.0 Pro (seedance-1.0-pro) Provider label: Seedance. Modes: text, image. Operation: generation. Prompt: required; maximum 2000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480p, 720p, 1080p. Durations: {"min":2,"max":12,"step":1}. Aspect ratios by mode: {"text":["21:9","16:9","4:3","1:1","3:4","9:16"],"image":["auto","21:9","16:9","4:3","1:1","3:4","9:16"]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"1080p","duration":5,"aspectRatio":"16:9","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: true. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Wan 2.5 (wan-2.5) Provider label: Wan. Modes: text, image. Operation: generation. Prompt: required; maximum 1500 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":1,"total":1},"image":{"images":1,"videos":0,"audio":1,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480p, 720p, 1080p. Durations: [5,10]. Aspect ratios by mode: {"text":["16:9","9:16","1:1"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"1080p","duration":5,"aspectRatio":"16:9","audio":true}. Audio: native. End image: false. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Kling 2.6 Pro (kling-2.6-pro) Provider label: Kling. Modes: text, image. Operation: generation. Prompt: required; maximum 2500 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 1080p. Durations: [5,10]. Aspect ratios by mode: {"text":["16:9","9:16","1:1"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"1080p","duration":5,"aspectRatio":"16:9","audio":true}. Audio: optional. End image: true. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Veo 3.1 (veo-3.1) Provider label: Google. Modes: text, image. Operation: generation. Prompt: required; maximum 20000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":1,"videos":0,"audio":0,"total":1}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 720p, 1080p, 4k. Durations: [4,6,8]. Aspect ratios by mode: {"text":["16:9","9:16"],"image":["auto","16:9","9:16"]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"720p","duration":8,"aspectRatio":"16:9","audio":true}. Audio: optional. End image: false. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### HappyHorse 1.0 (happyhorse-1.0) Provider label: Alibaba. Modes: text, image. Operation: generation. Prompt: required; maximum 2500 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":1,"videos":0,"audio":0,"total":1}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 720p, 1080p. Durations: {"min":3,"max":15,"step":1}. Aspect ratios by mode: {"text":["16:9","9:16","1:1","4:3","3:4"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"1080p","duration":5,"aspectRatio":"16:9","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### HappyHorse 1.1 (happyhorse-1.1) Provider label: Alibaba. Modes: text, image. Operation: generation. Prompt: required; maximum 2500 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":1,"videos":0,"audio":0,"total":1}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 720p, 1080p. Durations: {"min":3,"max":15,"step":1}. Aspect ratios by mode: {"text":["16:9","9:16","1:1","4:3","3:4","21:9","9:21","5:4","4:5"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"1080p","duration":5,"aspectRatio":"16:9","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Grok Imagine 1.5 (grok-imagine-1.5) Provider label: Grok. Modes: text, image. Operation: generation. Prompt: required; maximum 4096 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":1,"videos":0,"audio":0,"total":1}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480p, 720p, 1080p. Durations: {"min":1,"max":15,"step":1}. Aspect ratios by mode: {"text":["16:9","4:3","3:2","1:1","2:3","3:4","9:16"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"720p","duration":6,"aspectRatio":"16:9","audio":true}. Audio: native. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### H3 Max Styles (h3-max-styles) Provider label: MiniMax. Modes: text, image. Operation: generation. Prompt: required; maximum 49868 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":1,"videos":0,"audio":0,"total":1}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 768P. Durations: {"min":5,"max":15,"step":1}. Aspect ratios by mode: {"text":["21:9","16:9","4:3","1:1","3:4","9:16"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"768P","duration":5,"aspectRatio":"16:9","audio":true}. Audio: native. End image: false. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: ["hand-drawn","vhs","retro-toon-70s","low-poly","16bit-pixel"]. Output formats: ["mp4"]. ### H3 Max Camera Control (h3-max-camera-control) Provider label: MiniMax. Modes: image. Operation: generation. Prompt: optional; maximum 50000 characters. Input maxima by mode: {"image":{"images":1,"videos":0,"audio":0,"total":1}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480P, 768P, 1080P. Durations: {"min":3,"max":15,"step":1}. Aspect ratios by mode: {"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"image","resolution":"768P","duration":5,"aspectRatio":"16:9","audio":false}. Audio: no generated-audio switch; operation may retain source audio. End image: false. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: ["orbit-right","orbit-left","push-in","pull-back","crane-up"]. Output formats: ["mp4"]. ### H3 Max Lip Sync (h3-max-lip-sync) Provider label: MiniMax. Modes: reference. Operation: h3-lipsync. Prompt: not used; maximum 2000 characters. Input maxima by mode: {"reference":{"images":1,"videos":0,"audio":1,"total":12}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480P, 768P, 1080P, 2K. Durations: [5]. Aspect ratios by mode: {"reference":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"reference","resolution":"768P","duration":5,"aspectRatio":"adaptive","audio":true}. Audio: native. End image: false. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### Gemini Omni (gemini-omni) Provider label: Google. Modes: text, image. Operation: generation. Prompt: required; maximum 20000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":1,"videos":0,"audio":0,"total":1}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 720p. Durations: {"min":3,"max":10,"step":1}. Aspect ratios by mode: {"text":["16:9","9:16"],"image":["16:9","9:16"]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"720p","duration":8,"aspectRatio":"16:9","audio":true}. Audio: native. End image: false. Seed: false. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ### H3 LoRA (h3-lora) Provider label: MiniMax. Modes: text, image. Operation: generation. Prompt: required; maximum 50000 characters. Input maxima by mode: {"text":{"images":0,"videos":0,"audio":0,"total":0},"image":{"images":2,"videos":0,"audio":0,"total":2}}. Text mode does not require an image; image mode requires a start image and accepts an optional end image only if supported. Reference modes require their operation-specific inputs. Resolutions: 480P, 768P, 2K, 4K. Durations: {"min":5,"max":15,"step":1}. Aspect ratios by mode: {"text":["21:9","16:9","4:3","1:1","3:4","9:16"],"image":[]}; an empty list means framing is derived from the source or fixed, not arbitrary ratios. Defaults: {"mode":"text","resolution":"768P","duration":5,"aspectRatio":"16:9","audio":true}. Audio: native. End image: true. Seed: true. Web search: false. Last frame: false. Versions: []. Styles: []. Output formats: ["mp4"]. ## Neon AI-specific input rules Motion Transfer and Object Swap require one MP4/MOV source of at least four seconds, selecting/exporting a 4–30 second segment (otherwise first 30 seconds), and 1–8 ordered reference images. Object Swap also requires at least 409,600 pixels per frame. Restyle requires one source and a currently valid style from the live style catalog, with 0–5 character images. Prompts are optional. Outputs are video; do not assume source frame count, exact cadence or output dimensions are identical. Presets are demonstrations with source attribution, not proof of generations made by this site. Source file limit: 200 MiB; each reference image limit: 64 MiB. ## Currently enabled image integrations ### Seedream 5.0 Pro (seedream-5-pro) Provider label: Seedream. Modes: layer-decomposition, text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 5000 characters. Maximum reference images: 10. Accepted image types: ["workspace-supported image types"]. Maximum bytes: 31457280. Resolutions: ["1K","2K"]; default 2K. Ratios: ["1:1","16:9","9:16","4:3","3:4","3:2","2:3"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is layer-decomposition. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: true; maximum 6. Layer resolutions: ["auto","1K","1.5K","2K"]. ### GPT Image 2.6 Flare (gpt-image-2.6-flare) Provider label: OpenAI. Modes: text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 32000 characters. Maximum reference images: 16. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: ["1K","2K","4K"]; default 1K. Ratios: ["1:1","auto","3:2","2:3","4:3","3:4","16:9","9:16","2:1","1:2","3:1","1:3","21:9","9:21","5:4","4:5"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: true; maximum 10. Layer resolutions: []. Additional quality/background/format/mask/size rules: {"prompt":{"minLength":1,"maxLength":32000},"image_size":{"presets":["auto","square_hd","square","portrait_4_3","portrait_16_9","landscape_4_3","landscape_16_9"],"multiple":16,"maxEdge":3840,"minPixels":655360,"maxPixels":8294400,"maxAspectRatio":3},"quality":{"values":["auto","low","medium","high","xhigh","max"],"default":"high"},"background":{"values":["auto","transparent","opaque"],"default":"auto"},"num_images":{"min":1,"max":10,"default":1},"output_format":{"values":["png","jpeg","webp","gif","svg"],"default":"png"},"output_compression":{"min":0,"max":100,"formats":["jpeg","webp"]},"image_urls":{"modes":["image"],"max":16},"mask_url":{"modes":["image"],"optional":true},"sync_mode":{"value":false}}. ### GPT Image 2.6 Sunburst (gpt-image-2.6-sunburst) Provider label: OpenAI. Modes: text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 32000 characters. Maximum reference images: 16. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: ["1K","2K","4K"]; default 1K. Ratios: ["1:1","auto","3:2","2:3","4:3","3:4","16:9","9:16","2:1","1:2","3:1","1:3","21:9","9:21","5:4","4:5"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: true; maximum 10. Layer resolutions: []. Additional quality/background/format/mask/size rules: {"prompt":{"minLength":1,"maxLength":32000},"image_size":{"presets":["auto","square_hd","square","portrait_4_3","portrait_16_9","landscape_4_3","landscape_16_9"],"multiple":16,"maxEdge":3840,"minPixels":655360,"maxPixels":8294400,"maxAspectRatio":3},"quality":{"values":["auto","low","medium","high","xhigh","max"],"default":"high"},"background":{"values":["auto","transparent","opaque"],"default":"auto"},"num_images":{"min":1,"max":10,"default":1},"output_format":{"values":["png","jpeg","webp","gif","svg"],"default":"png"},"output_compression":{"min":0,"max":100,"formats":["jpeg","webp"]},"image_urls":{"modes":["image"],"max":16},"mask_url":{"modes":["image"],"optional":true},"sync_mode":{"value":false}}. ### GPT Image 2.5 Flare (gpt-image-2.5-flare) Provider label: OpenAI. Modes: text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 32000 characters. Maximum reference images: 16. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: ["1K","2K","4K"]; default 1K. Ratios: ["1:1","auto","3:2","2:3","4:3","3:4","16:9","9:16","2:1","1:2","3:1","1:3","21:9","9:21","5:4","4:5"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: true; maximum 10. Layer resolutions: []. Additional quality/background/format/mask/size rules: {"prompt":{"minLength":1,"maxLength":32000},"image_size":{"presets":["auto","square_hd","square","portrait_4_3","portrait_16_9","landscape_4_3","landscape_16_9"],"multiple":16,"maxEdge":3840,"minPixels":655360,"maxPixels":8294400,"maxAspectRatio":3},"quality":{"values":["auto","low","medium","high","xhigh","max"],"default":"high"},"background":{"values":["auto","transparent","opaque"],"default":"auto"},"num_images":{"min":1,"max":10,"default":1},"output_format":{"values":["png","jpeg","webp"],"default":"png"},"output_compression":{"min":0,"max":100,"formats":["jpeg","webp"]},"image_urls":{"modes":["image"],"max":16},"mask_url":{"modes":["image"],"optional":true},"sync_mode":{"value":false}}. ### GPT Image 2.5 Sunburst (gpt-image-2.5-sunburst) Provider label: OpenAI. Modes: text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 32000 characters. Maximum reference images: 16. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: ["1K","2K","4K"]; default 1K. Ratios: ["1:1","auto","3:2","2:3","4:3","3:4","16:9","9:16","2:1","1:2","3:1","1:3","21:9","9:21","5:4","4:5"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: true; maximum 10. Layer resolutions: []. Additional quality/background/format/mask/size rules: {"prompt":{"minLength":1,"maxLength":32000},"image_size":{"presets":["auto","square_hd","square","portrait_4_3","portrait_16_9","landscape_4_3","landscape_16_9"],"multiple":16,"maxEdge":3840,"minPixels":655360,"maxPixels":8294400,"maxAspectRatio":3},"quality":{"values":["auto","low","medium","high","xhigh","max"],"default":"high"},"background":{"values":["auto","transparent","opaque"],"default":"auto"},"num_images":{"min":1,"max":10,"default":1},"output_format":{"values":["png","jpeg","webp"],"default":"png"},"output_compression":{"min":0,"max":100,"formats":["jpeg","webp"]},"image_urls":{"modes":["image"],"max":16},"mask_url":{"modes":["image"],"optional":true},"sync_mode":{"value":false}}. ### GPT Image 2 (gpt-image-2) Provider label: OpenAI. Modes: text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 20000 characters. Maximum reference images: 16. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: ["1K","2K","4K"]; default 1K. Ratios: ["auto","1:1","3:2","2:3","4:3","3:4","16:9","9:16","2:1","1:2","3:1","1:3","21:9","9:21","5:4","4:5"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: true; maximum 4. Layer resolutions: []. ### Nano Banana 2 (nano-banana-2) Provider label: Nano Banana. Modes: text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 20000 characters. Maximum reference images: 14. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: ["1K","2K","4K"]; default 1K. Ratios: ["auto","1:1","2:3","3:2","3:4","4:3","4:5","5:4","9:16","16:9","21:9"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: true; maximum 4. Layer resolutions: []. ### Nano Banana Pro (nano-banana-pro) Provider label: Nano Banana. Modes: text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 20000 characters. Maximum reference images: 14. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: ["1K","2K","4K"]; default 1K. Ratios: ["1:1","2:3","3:2","3:4","4:3","4:5","5:4","9:16","16:9","21:9"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: true; maximum 4. Layer resolutions: []. ### Nano Banana (nano-banana) Provider label: Nano Banana. Modes: text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 20000 characters. Maximum reference images: 10. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: []; default model default. Ratios: ["1:1","16:9","9:16","4:3","3:4","3:2","2:3","5:4","4:5","21:9"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: true; maximum 4. Layer resolutions: []. ### Seedream 5 Lite (seedream-5-lite) Provider label: Seedream. Modes: text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 5000 characters. Maximum reference images: 10. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: ["2K","3K"]; default 2K. Ratios: ["1:1","16:9","9:16","4:3","3:4","3:2","2:3","5:4","4:5","21:9"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: true; maximum 4. Layer resolutions: []. ### Seedream 4.5 (seedream-4-5) Provider label: Seedream. Modes: text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 3000 characters. Maximum reference images: 10. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: ["2K","4K"]; default 2K. Ratios: ["1:1","16:9","9:16","4:3","3:4","3:2","2:3","5:4","4:5","21:9"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: true; maximum 4. Layer resolutions: []. ### Seedream V4 (seedream-v4) Provider label: Seedream. Modes: text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 5000 characters. Maximum reference images: 10. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: ["1K","2K","4K"]; default 1K. Ratios: ["1:1","16:9","9:16","4:3","3:4","3:2","2:3","5:4","4:5","21:9"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: true; maximum 4. Layer resolutions: []. ### Bria RMBG 2.0 (bria-background-remove) Provider label: Bria. Modes: image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: false; maximum 0 characters. Maximum reference images: 1. Accepted image types: ["image/jpeg","image/png","image/webp"]. Maximum bytes: 31457280. Resolutions: ["original"]; default original. Ratios: ["auto"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is image. Output format: png. Multiple outputs: false; maximum 1. Layer resolutions: []. ### Illusion Diffusion (illusion-diffusion) Provider label: Illusion Diffusion. Modes: image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 2000 characters. Maximum reference images: 1. Accepted image types: ["image/jpeg","image/png","image/webp"]. Maximum bytes: 10485760. Resolutions: ["1K"]; default 1K. Ratios: ["1:1"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is image. Output format: png. Multiple outputs: false; maximum 1. Layer resolutions: []. ### Z-Image (z-image) Provider label: Z-Image. Modes: text. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 1000 characters. Maximum reference images: 0. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: []; default model default. Ratios: ["1:1","4:3","3:4","16:9","9:16"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: workspace PNG/JPEG/WebP choice. Multiple outputs: false; maximum 1. Layer resolutions: []. ### Nano Banana 2 Lite (nano-banana-2-lite) Provider label: Nano Banana. Modes: text. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 50000 characters. Maximum reference images: 0. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: ["1K"]; default 1K. Ratios: ["auto","1:1","21:9","16:9","3:2","4:3","5:4","4:5","3:4","2:3","9:16","4:1","1:4","8:1","1:8"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: png. Multiple outputs: true; maximum 4. Layer resolutions: []. ### GPT Image 1.5 (gpt-image-1.5) Provider label: OpenAI. Modes: text, image. Text generation requires a prompt unless configured otherwise; image editing requires reference image input. Prompt required: true; maximum 32000 characters. Maximum reference images: 10. Accepted image types: ["workspace-supported image types"]. Maximum bytes: workspace upload validation. Resolutions: ["1K"]; default 1K. Ratios: ["1:1","3:2","2:3"]. Modes start from the page-specific defaults or restored model draft; the catalog's first mode is text. Output format: png. Multiple outputs: true; maximum 4. Layer resolutions: []. ## GPT Image 2.6 native formats GIF and SVG tools default to GPT Image 2.6 Flare. GIF uses seven frames at 200 ms and a total 8,294,400-pixel budget; SVG permits one output with GPT-6 Astra vector conversion. Quotes follow the inherited output-token estimate rule at $30/million image output tokens, USD × 100 rounded up with a four-credit/image minimum. Input, reasoning and SVG conversion provider charges are not claimed as exact output estimates. Review the workspace quote. ## Separate workflows ### Anthropic Opus 5.5 code rendering Image ID claude-opus-5-5-image; video ID claude-opus-5-5-motion; provider anthropic/claude-opus-5.5 via fal. Input $4/million tokens; output $20/million tokens. Validated scene instructions render to PNG or silent MP4 in the browser; no arbitrary code executes. ### OpenAI GPT-6 Astra code rendering Image ID gpt-6-astra-image; video ID gpt-6-astra-motion; provider openai/gpt-6-astra via fal. Input $10/million tokens; output $50/million tokens. Validated scene instructions render to PNG or silent MP4 in the browser; no arbitrary code executes. ### Google Gemini 4 Argon code rendering Image ID gemini-4-argon-image; video ID gemini-4-argon-motion; provider google/gemini-4-argon via fal. Input $2/million tokens; output $10/million tokens. Validated scene instructions render to PNG or silent MP4 in the browser; no arbitrary code executes. ### Opus motion graphics (claude-opus-5-5-motion) {"id":"claude-opus-5-5-motion","name":"Opus 5.5 · Code Video","provider":"Anthropic","modes":["text"],"resolutions":[{"id":"720p","label":"720p"},{"id":"1080p","label":"1080p"}],"durations":[10,15,20,30],"aspectRatios":{"text":[{"id":"16:9","label":"16:9"},{"id":"9:16","label":"9:16"},{"id":"1:1","label":"1:1"}]},"defaults":{"mode":"text","resolution":"720p","duration":20,"aspectRatio":"16:9","audio":false},"uploadLimits":{"images":4,"videos":0,"audio":0},"styles":[{"id":"sketch","label":"Sketch"},{"id":"motion","label":"Motion"}]} An instruction or confirmed website brief plus approved image elements drives a validated editable scene. Create/edit and browser MP4 export are separate steps, not ordinary pixel-video generation. https://neonai.io/opus-5-5-video-generator. ### Humanize Model google/gemini-2.5-flash; up to 12000 source characters, tones natural, friendly, professional; outputs rewritten text. https://neonai.io/humanize-ai. ### Prompt Seen Model google/gemini-2.5-flash; requires a normalized JPEG up to 1024 pixels/side and optional direction up to 1000 characters; tones detailed, concise, cinematic; outputs a text prompt, not an image. https://neonai.io/prompt-seen. ### Layer decomposition Seedream 5 Pro separates one input image into layers; supported layer resolutions come from its model configuration above. Output count is discovered after generation; pricing.md explains reservation and excess-credit release. ### H3 LoRA inference The H3 LoRA integration requires a user-supplied compatible LoRA identifier or URL and supports a strength setting. This is inference with supplied weights, not a hosted training service. Use the model settings and quote in the workspace. ## Code integrations currently unavailable - Opus 5.5 · Code Image (claude-opus-5-5-image): required provider credential not configured. - GPT-6 Astra · Code Image (gpt-6-astra-image): required provider credential not configured. - Gemini 4 Argon · Code Image (gemini-4-argon-image): required provider credential not configured. ## Legacy AI playground integrations Workspace: https://neonai.io/ai. These are separate demo operations, not additional selectable studio models. Workers AI configuration: unavailable. Operations and model IDs: {"summarization":"@cf/meta/llama-3.2-3b-instruct","tagline":"@cf/meta/llama-3.2-3b-instruct","translation":"@cf/meta/m2m100-1.2b","textToSpeech":"@cf/deepgram/aura-1","imageCaption":"@cf/llava-hf/llava-1.5-7b-hf"}. Text input produces summaries, translations, taglines or speech; captioning accepts an image and a question. Workers AI text-to-image choices: ["@cf/black-forest-labs/flux-1-schnell","@cf/bytedance/stable-diffusion-xl-lightning","@cf/lykon/dreamshaper-8-lcm"]. Fal demo configuration: configured. Text-to-image choices: ["fal-ai/flux/schnell","fal-ai/nano-banana","openai/gpt-image-2"]. Image-edit integration: fal-ai/nano-banana/edit; image plus prompt produces an edited image. These demo functions do not use the main studio reservation/settlement flow; the one-credit UI threshold is not a verified per-call billed price. ## Local browser tools (not generative models) [Compress Image](https://neonai.io/compress-image) [Resize Image](https://neonai.io/resize-image) [Convert Image](https://neonai.io/convert-image) [Rotate Image](https://neonai.io/rotate-image) [Crop Image](https://neonai.io/crop-image) [Watermark Image](https://neonai.io/watermark-image) [Background Remover](https://neonai.io/background-remover) [Transparent Background](https://neonai.io/transparent-background) [White Background](https://neonai.io/white-background) [Change Background Color](https://neonai.io/change-background-color) These process images locally, require login to process/save and retain their existing free-download behavior. They are distinct from paid AI background removal/editing models. No training service or unspecified model behind a competitor-named page is promised by this catalog.