seedance-2.5); a mode is what it does with your inputs (text-to-video, image-to-video, reference-to-video…). Each model page lists, per mode, the fields it takes, their allowed values, the price and an example.
Your assistant reads the same catalog with search_models and get_model, so it always has the current fields and prices. The catalog is per workspace: it matches the models enabled in your playground.
Prices are in credits. ¹ Not enabled by default: a workspace admin turns it on in Argil, under Settings › Connectors.
Video
| Model | Mode | Price |
|---|---|---|
seedance-2.5 | text-to-video | ~11.18 credits per second at 720p (Seedance 2.5; rate varies by resolution; output + input-video seconds both billed) |
seedance-2.5 | image-to-video | ~11.18 credits per second at 720p (Seedance 2.5; rate varies by resolution; output + input-video seconds both billed) |
seedance-2.5 | reference-to-video | ~11.18 credits per second at 720p (Seedance 2.5; rate varies by resolution; output + input-video seconds both billed) |
seedance-2.5 | extend | ~11.18 credits per second at 720p (Seedance 2.5; rate varies by resolution; output + input-video seconds both billed) |
seedance-2.5 | edit | ~11.18 credits per second at 720p (Seedance 2.5; rate varies by resolution; output + input-video seconds both billed) |
minimax-h3-max | text-to-video | ~2.42 credits/s at 480P (fastest); 768P is $0.08/s |
minimax-h3-max | image-to-video | ~2.42 credits/s at 480P (fastest); 768P is $0.08/s |
minimax-h3-max | reference-to-video | 2.42/3.87/7.74 credits/s at 480P/768P/1080P; 480P default, plus estimated reference charges. |
flux-3-max | text-to-video | ~8.23 credits/s at 720p; 1080p is $0.29/s |
flux-3-max | image-to-video | ~8.23 credits/s at 720p; 1080p is $0.29/s |
ltx-2.3 ¹ | text-to-video | 2.90 credits per second (1080p, with audio) |
ltx-2.3 ¹ | image-to-video | 2.90 credits per second (1080p, with audio) |
ltx-2.3 ¹ | audio-to-video | 4.84 credits per second of input audio (the video matches its length; 10s assumed when unknown) |
kling-2.6-pro-motion-control ¹ | motion-transfer | 5.42 credits per second of input video |
seedance-2.0 | text-to-video | ~7.32 credits per second at 720p (Seedance 2.0; rate varies by resolution; output + input-video seconds both billed) |
seedance-2.0 | image-to-video | ~7.32 credits per second at 720p (Seedance 2.0; rate varies by resolution; output + input-video seconds both billed) |
seedance-2.0 | reference-to-video | ~7.32 credits per second at 720p (Seedance 2.0; rate varies by resolution; output + input-video seconds both billed) |
seedance-2.0-mini | text-to-video | ~3.66 credits per second at 720p (Seedance 2.0 Mini; rate varies by resolution; output + input-video seconds both billed) |
seedance-2.0-mini | image-to-video | ~3.66 credits per second at 720p (Seedance 2.0 Mini; rate varies by resolution; output + input-video seconds both billed) |
seedance-2.0-mini | reference-to-video | ~3.66 credits per second at 720p (Seedance 2.0 Mini; rate varies by resolution; output + input-video seconds both billed) |
pi-video ¹ | text-to-video | 0.97 credits/s (720p), 1.94 credits/s (1080p), draft: 0.24 credits/s (720p), 0.48 credits/s (1080p), per second of output video (audio-to-video: the audio’s length); settled on the seconds Replicate bills |
pi-video ¹ | image-to-video | 0.97 credits/s (720p), 1.94 credits/s (1080p), draft: 0.24 credits/s (720p), 0.48 credits/s (1080p), per second of output video (audio-to-video: the audio’s length); settled on the seconds Replicate bills |
pi-video ¹ | audio-to-video | 0.97 credits/s (720p), 1.94 credits/s (1080p), draft: 0.24 credits/s (720p), 0.48 credits/s (1080p), per second of output video (audio-to-video: the audio’s length); settled on the seconds Replicate bills |
grok-imagine-video-1.5 | text-to-video | 3.87 credits/s (480p), 6.77 credits/s (720p), + 0.48 credits per input image (frame or reference) |
grok-imagine-video-1.5 | image-to-video | 3.87 credits/s (480p), 6.77 credits/s (720p), + 0.48 credits per input image (frame or reference) |
grok-imagine-video-1.5 | reference-to-video | 3.87 credits/s (480p), 6.77 credits/s (720p), + 0.48 credits per input image (frame or reference) |
minimax-h3 | text-to-video | ~3.87 credits/s at 768P (default; 2K ~6.29/s), output + reference-video input, first 5 reference images free then ~1.94 each. 480P (MiniMax H3 Max): ~2.42/s output, ~2.68/s reference video, first 2 images free then ~3.58 each. Audio free |
minimax-h3 | image-to-video | ~3.87 credits/s at 768P (default; 2K ~6.29/s), output + reference-video input, first 5 reference images free then ~1.94 each. 480P (MiniMax H3 Max): ~2.42/s output, ~2.68/s reference video, first 2 images free then ~3.58 each. Audio free |
minimax-h3 | reference-to-video | ~3.87 credits/s at 768P (default; 2K ~6.29/s), output + reference-video input, first 5 reference images free then ~1.94 each. 480P (MiniMax H3 Max): ~2.42/s output, ~2.68/s reference video, first 2 images free then ~3.58 each. Audio free |
argil-atom | script-to-video | TTS at the voice model’s rate (free on your own key) + Atom (150 cr/min audio) |
argil-atom | audio-to-video | 150 credits/min of audio (Argil Atom lip-sync) |
Image
| Model | Mode | Price |
|---|---|---|
nano-banana-2.1 | text-to-image | 1.94–6.48 credits per image by resolution |
nano-banana-2.1 | image-to-image | 1.94–6.48 credits per image by resolution |
grok-imagine-image-2.0 | text-to-image | 1.94–3.87 credits per image by quality and resolution (generate defaults to low, edit to medium), + 0.48 per input image |
grok-imagine-image-2.0 | image-to-image | 1.94–3.87 credits per image by quality and resolution (generate defaults to low, edit to medium), + 0.48 per input image |
grok-imagine-image-2.0 | reference-to-image | 1.94–3.87 credits per image by quality and resolution (generate defaults to low, edit to medium), + 0.48 per input image |
gpt-image-2.5-flare | text-to-image | fal’s token bill per image, by size × quality. High: 1K 2.55 (1:1) / 1.75 (4:3) / 1.25 (16:9), minimum 1.37–2.23, 2K 2.05–5.18, 4K 4.84–8.61 credits; low ≈ 0.11× and medium ≈ 0.25× of high. Default (high, no size) 1.75. Each reference image adds 0.41; prompt text adds $5/M tokens. |
gpt-image-2.5-flare | image-to-image | fal’s token bill per image, by size × quality. High: 1K 2.55 (1:1) / 1.75 (4:3) / 1.25 (16:9), minimum 1.37–2.23, 2K 2.05–5.18, 4K 4.84–8.61 credits; low ≈ 0.11× and medium ≈ 0.25× of high. Default (high, no size) 2.55. Each reference image adds 0.41; prompt text adds $5/M tokens. |
gpt-image-2.5-sunburst | text-to-image | fal’s token bill per image, by size × quality. High: 1K 2.55 (1:1) / 1.75 (4:3) / 1.25 (16:9), minimum 1.37–2.23, 2K 2.05–5.18, 4K 4.84–8.61 credits; low ≈ 0.11× and medium ≈ 0.25× of high. Default (high, no size) 1.75. Each reference image adds 0.41; prompt text adds $5/M tokens. |
gpt-image-2.5-sunburst | image-to-image | fal’s token bill per image, by size × quality. High: 1K 2.55 (1:1) / 1.75 (4:3) / 1.25 (16:9), minimum 1.37–2.23, 2K 2.05–5.18, 4K 4.84–8.61 credits; low ≈ 0.11× and medium ≈ 0.25× of high. Default (high, no size) 2.55. Each reference image adds 0.41; prompt text adds $5/M tokens. |
gpt-image-2.6-flare | text-to-image | fal’s token bill per image, by size × quality. High: 1K 2.55 (1:1) / 1.75 (4:3) / 1.25 (16:9), minimum 1.37–2.23, 2K 2.05–5.18, 4K 4.84–8.61 credits; low ≈ 0.11× and medium ≈ 0.25× of high. Default (high, no size) 1.75. Each reference image adds 0.41; prompt text adds $5/M tokens. |
gpt-image-2.6-flare | image-to-image | fal’s token bill per image, by size × quality. High: 1K 2.55 (1:1) / 1.75 (4:3) / 1.25 (16:9), minimum 1.37–2.23, 2K 2.05–5.18, 4K 4.84–8.61 credits; low ≈ 0.11× and medium ≈ 0.25× of high. Default (high, no size) 2.55. Each reference image adds 0.41; prompt text adds $5/M tokens. |
gpt-image-2.6-sunburst | text-to-image | fal’s token bill per image, by size × quality. High: 1K 2.55 (1:1) / 1.75 (4:3) / 1.25 (16:9), minimum 1.37–2.23, 2K 2.05–5.18, 4K 4.84–8.61 credits; low ≈ 0.11× and medium ≈ 0.25× of high. Default (high, no size) 1.75. Each reference image adds 0.41; prompt text adds $5/M tokens. |
gpt-image-2.6-sunburst | image-to-image | fal’s token bill per image, by size × quality. High: 1K 2.55 (1:1) / 1.75 (4:3) / 1.25 (16:9), minimum 1.37–2.23, 2K 2.05–5.18, 4K 4.84–8.61 credits; low ≈ 0.11× and medium ≈ 0.25× of high. Default (high, no size) 2.55. Each reference image adds 0.41; prompt text adds $5/M tokens. |
seedream-5.0-pro | text-to-image | 3.27–6.53 credits/image; default 6.53 at auto_2K. |
seedream-5.0-pro | image-to-image | 3.27–6.53 credits/image; default 6.53 at auto_2K. Editing: +0.22 per additional reference after the first, per output image. |
ideogram-3.0-character | reference-to-image | 4.84–9.68 credits per image (by rendering speed; BALANCED default) |
flux-1-schnell ¹ | text-to-image | 0.1452 credits per megapixel |
flux-1-schnell ¹ | image-to-image | 1.2097 credits per megapixel |
flux-2-klein-4b | text-to-image | 0.4355 credits per megapixel |
flux-2-klein-4b | image-to-image | 0.4839 credits per megapixel |
nano-banana-2 | text-to-image | 2.90–7.74 credits per image by resolution (1K default: 3.87) |
nano-banana-2 | image-to-image | 2.90–7.74 credits per image by resolution (1K default: 3.87) |
Audio
| Model | Mode | Price |
|---|---|---|
lyria-2 ¹ | text-to-music | 3.87 credits per generated clip |
lyria-3-pro ¹ | text-to-music | 3.87 credits per generated clip |
elevenlabs-multilingual-2 | text-to-speech | 3.87 credits per 1k characters |
elevenlabs-multilingual-2 | speech-to-speech | 0.10 credits per second |
elevenlabs-3 | text-to-speech | 3.87 credits per 1k characters |
elevenlabs-4 | text-to-speech | 3.87 credits per 1k characters |
elevenlabs-turbo-2.5 | text-to-speech | 1.94 credits per 1k characters |
elevenlabs-english-2 | speech-to-speech | 0.10 credits per second |
minimax-speech-2.0-turbo | text-to-speech | 2.90 credits per 1k characters |
minimax-speech-2.0-hd | text-to-speech | 4.84 credits per 1k characters |
minimax-speech-2.5-turbo | text-to-speech | 1.94 credits per 1k characters |
minimax-speech-2.5-hd | text-to-speech | 2.90 credits per 1k characters |
Model lifecycle
- Models and modes are added at any time; their fields and prices can change as the model evolves.
- A change that would break existing requests ships as a new model id (
nano-banana-2.1→nano-banana-3). - A model is marked deprecated at least 90 days before it is removed, with a suggested replacement.