fal-endpoint-types 1.3.34 → 1.3.35

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -91,11 +91,11 @@ export interface XaiGrokImagineVideoTextToVideoOutput {
91
91
  * "height": 720,
92
92
  * "duration": 6.041667,
93
93
  * "url": "https://v3b.fal.media/files/b/0a8b90e4/RUAbFYlssdqnbjNLmE8qP_IX7BNYGP.mp4",
94
- * "width": 1280,
95
94
  * "fps": 24,
95
+ * "width": 1280,
96
96
  * "file_name": "RUAbFYlssdqnbjNLmE8qP_IX7BNYGP.mp4",
97
- * "content_type": "video/mp4",
98
- * "num_frames": 145
97
+ * "num_frames": 145,
98
+ * "content_type": "video/mp4"
99
99
  * }
100
100
  */
101
101
  video: Components.VideoFile;
@@ -142,11 +142,11 @@ export interface XaiGrokImagineVideoImageToVideoOutput {
142
142
  * "height": 720,
143
143
  * "duration": 6.041667,
144
144
  * "url": "https://v3b.fal.media/files/b/0a8b90e0/0Ci1dviuSnEyUZzBUq-_5_nu7MrAAa.mp4",
145
- * "width": 1280,
146
145
  * "fps": 24,
146
+ * "width": 1280,
147
147
  * "file_name": "0Ci1dviuSnEyUZzBUq-_5_nu7MrAAa.mp4",
148
- * "content_type": "video/mp4",
149
- * "num_frames": 145
148
+ * "num_frames": 145,
149
+ * "content_type": "video/mp4"
150
150
  * }
151
151
  */
152
152
  video: Components.VideoFile;
@@ -181,11 +181,11 @@ export interface XaiGrokImagineVideoEditVideoOutput {
181
181
  * "height": 720,
182
182
  * "duration": 5.041667,
183
183
  * "url": "https://v3b.fal.media/files/b/0a8b9113/EuDrZuQTW9m1phBXOsauz_EpJH3s8X.mp4",
184
- * "width": 1280,
185
184
  * "fps": 24,
185
+ * "width": 1280,
186
186
  * "file_name": "EuDrZuQTW9m1phBXOsauz_EpJH3s8X.mp4",
187
- * "content_type": "video/mp4",
188
- * "num_frames": 121
187
+ * "num_frames": 121,
188
+ * "content_type": "video/mp4"
189
189
  * }
190
190
  */
191
191
  video: Components.VideoFile;
@@ -13006,148 +13006,6 @@ export interface SharedType_151 {
13006
13006
  seed: number;
13007
13007
  }
13008
13008
 
13009
- export interface SharedType_133 {
13010
- /**
13011
- * Aspect Ratio
13012
- * @description The aspect ratio of the video.
13013
- * @default auto
13014
- * @example auto
13015
- * @enum {string}
13016
- */
13017
- aspect_ratio?: '9:16' | '1:1' | '16:9' | 'auto';
13018
- /**
13019
- * Constant Rate Factor
13020
- * @description The constant rate factor (CRF) to compress input media with. Compressed input media more closely matches the model's training data, which can improve motion quality.
13021
- * @default 29
13022
- * @example 29
13023
- */
13024
- constant_rate_factor?: number;
13025
- /**
13026
- * Enable Detail Pass
13027
- * @description Whether to use a detail pass. If True, the model will perform a second pass to refine the video and enhance details. This incurs a 2.0x cost multiplier on the base price.
13028
- * @default false
13029
- * @example false
13030
- */
13031
- enable_detail_pass?: boolean;
13032
- /**
13033
- * Enable Safety Checker
13034
- * @description Whether to enable the safety checker.
13035
- * @default true
13036
- * @example true
13037
- */
13038
- enable_safety_checker?: boolean;
13039
- /**
13040
- * Expand Prompt
13041
- * @description Whether to expand the prompt using a language model.
13042
- * @default false
13043
- * @example false
13044
- */
13045
- expand_prompt?: boolean;
13046
- /**
13047
- * Number of Inference Steps
13048
- * @description Number of inference steps during the first pass.
13049
- * @default 8
13050
- * @example 8
13051
- */
13052
- first_pass_num_inference_steps?: number;
13053
- /**
13054
- * Frame Rate
13055
- * @description The frame rate of the video.
13056
- * @default 24
13057
- * @example 24
13058
- */
13059
- frame_rate?: number;
13060
- /**
13061
- * Loras
13062
- * @description LoRA weights to use for generation
13063
- * @default []
13064
- */
13065
- loras?: Components.LoRAWeight[];
13066
- /**
13067
- * Negative Prompt
13068
- * @description Negative prompt for generation
13069
- * @default worst quality, inconsistent motion, blurry, jittery, distorted
13070
- */
13071
- negative_prompt?: string;
13072
- /**
13073
- * Number of Frames
13074
- * @description The number of frames in the video.
13075
- * @default 121
13076
- * @example 121
13077
- */
13078
- num_frames?: number;
13079
- /**
13080
- * Prompt
13081
- * @description Text prompt to guide generation
13082
- * @example Woman walking on a street in Tokyo
13083
- */
13084
- prompt: string;
13085
- /**
13086
- * Resolution
13087
- * @description Resolution of the generated video.
13088
- * @default 720p
13089
- * @example 720p
13090
- * @enum {string}
13091
- */
13092
- resolution?: '480p' | '720p';
13093
- /**
13094
- * Reverse Video
13095
- * @description Whether to reverse the video.
13096
- * @default false
13097
- * @example false
13098
- */
13099
- reverse_video?: boolean;
13100
- /**
13101
- * Second Pass Number of Inference Steps
13102
- * @description Number of inference steps during the second pass.
13103
- * @default 8
13104
- * @example 8
13105
- */
13106
- second_pass_num_inference_steps?: number;
13107
- /**
13108
- * Second Pass Skip Initial Steps
13109
- * @description The number of inference steps to skip in the initial steps of the second pass. By skipping some steps at the beginning, the second pass can focus on smaller details instead of larger changes.
13110
- * @default 5
13111
- * @example 5
13112
- */
13113
- second_pass_skip_initial_steps?: number;
13114
- /**
13115
- * Seed
13116
- * @description Random seed for generation
13117
- */
13118
- seed?: number;
13119
- /**
13120
- * Temporal AdaIN Factor
13121
- * @description The factor for adaptive instance normalization (AdaIN) applied to generated video chunks after the first. This can help deal with a gradual increase in saturation/contrast in the generated video by normalizing the color distribution across the video. A high value will ensure the color distribution is more consistent across the video, while a low value will allow for more variation in color distribution.
13122
- * @default 0.5
13123
- * @example 0.5
13124
- */
13125
- temporal_adain_factor?: number;
13126
- /**
13127
- * Tone Map Compression Ratio
13128
- * @description The compression ratio for tone mapping. This is used to compress the dynamic range of the video to improve visual quality. A value of 0.0 means no compression, while a value of 1.0 means maximum compression.
13129
- * @default 0
13130
- * @example 0
13131
- */
13132
- tone_map_compression_ratio?: number;
13133
- /**
13134
- * @description Video to be extended.
13135
- * @example {
13136
- * "video_url": "https://storage.googleapis.com/falserverless/web-examples/wan/t2v.mp4",
13137
- * "start_frame_num": 0,
13138
- * "reverse_video": false,
13139
- * "limit_num_frames": false,
13140
- * "resample_fps": false,
13141
- * "strength": 1,
13142
- * "target_fps": 24,
13143
- * "max_num_frames": 1441,
13144
- * "conditioning_type": "rgb",
13145
- * "preprocess": false
13146
- * }
13147
- */
13148
- video: Components.ExtendVideoConditioningInput;
13149
- }
13150
-
13151
13009
  export interface SharedType_129 {
13152
13010
  /**
13153
13011
  * Results
@@ -15357,7 +15215,7 @@ export interface ZImageTurboTilingLoraInput {
15357
15215
  enable_safety_checker?: boolean;
15358
15216
  /**
15359
15217
  * Image Size
15360
- * @description The size of the generated image.
15218
+ * @description The size of the generated image. Use 'auto' to match the input image size (or 1024x1024 if no image).
15361
15219
  * @default square_hd
15362
15220
  */
15363
15221
  image_size?:
@@ -15369,7 +15227,13 @@ export interface ZImageTurboTilingLoraInput {
15369
15227
  | 'portrait_16_9'
15370
15228
  | 'landscape_4_3'
15371
15229
  | 'landscape_16_9'
15230
+ | 'auto'
15372
15231
  );
15232
+ /**
15233
+ * Image URL
15234
+ * @description URL of an image for image-to-image or inpainting. When provided without mask_image_url, performs image-to-image; with mask_image_url, performs inpainting.
15235
+ */
15236
+ image_url?: string;
15373
15237
  /**
15374
15238
  * Loras
15375
15239
  * @description List of LoRA weights to apply (maximum 3).
@@ -15377,100 +15241,10 @@ export interface ZImageTurboTilingLoraInput {
15377
15241
  */
15378
15242
  loras?: Components.LoRAInput_2[];
15379
15243
  /**
15380
- * Number of Images
15381
- * @description The number of images to generate.
15382
- * @default 1
15383
- */
15384
- num_images?: number;
15385
- /**
15386
- * Number of Inference Steps
15387
- * @description The number of inference steps to perform.
15388
- * @default 8
15389
- */
15390
- num_inference_steps?: number;
15391
- /**
15392
- * Output Format
15393
- * @description The format of the generated image.
15394
- * @default png
15395
- * @enum {string}
15396
- */
15397
- output_format?: 'jpeg' | 'png' | 'webp';
15398
- /**
15399
- * Prompt
15400
- * @description The prompt to generate an image from.
15401
- * @example A hyper-realistic, high-resolution 4k texture of an ancient weathered brick wall heavily overgrown with lush green moss and soft lichens. The bricks are aged, featuring deep earthy tones, natural cracks, and gritty textures. Vibrant emerald moss fills the mortar lines and spills over the rough surfaces of the stones. Uniform, flat cinematic lighting ensures no harsh shadows, highlighting the intricate organic details and damp stone surfaces. The composition is a perfectly balanced overhead view, showcasing a rich tapestry of botanical growth and masonry craftsmanship with professional clarity and hyper-detailed grit.
15402
- */
15403
- prompt: string;
15404
- /**
15405
- * Seed
15406
- * @description The same seed and the same prompt given to the same version of the model
15407
- * will output the same image every time.
15408
- */
15409
- seed?: number;
15410
- /**
15411
- * Sync Mode
15412
- * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
15413
- * @default false
15414
- */
15415
- sync_mode?: boolean;
15416
- /**
15417
- * Tile Size
15418
- * @description Tile size in latent space (64 = 512px, 128 = 1024px, 256 = 2048px).
15419
- * @default 128
15420
- */
15421
- tile_size?: number;
15422
- /**
15423
- * Tile Stride
15424
- * @description Tile stride in latent space. (32 = 256px, 64 = 512px, 128 = 1024px).
15425
- * @default 64
15426
- */
15427
- tile_stride?: number;
15428
- /**
15429
- * Tiling Mode
15430
- * @description Tiling direction: 'both' (omnidirectional), 'horizontal' (left-right only), 'vertical' (top-bottom only).
15431
- * @default both
15432
- * @enum {string}
15433
- */
15434
- tiling_mode?: 'both' | 'horizontal' | 'vertical';
15435
- }
15436
-
15437
- export interface ZImageTurboTilingLoraOutput extends SharedType_390 {}
15438
-
15439
- export interface ZImageTurboTilingInput {
15440
- /**
15441
- * Acceleration
15442
- * @description The acceleration level to use.
15443
- * @default regular
15444
- * @enum {string}
15445
- */
15446
- acceleration?: 'none' | 'regular' | 'high';
15447
- /**
15448
- * Enable Prompt Expansion
15449
- * @description Whether to enable prompt expansion. Note: this will increase the price by 0.0025 credits per request.
15450
- * @default false
15451
- */
15452
- enable_prompt_expansion?: boolean;
15453
- /**
15454
- * Enable Safety Checker
15455
- * @description If set to true, the safety checker will be enabled.
15456
- * @default true
15457
- */
15458
- enable_safety_checker?: boolean;
15459
- /**
15460
- * Image Size
15461
- * @description The size of the generated image.
15462
- * @default square_hd
15244
+ * Mask Image URL
15245
+ * @description URL of a mask image for inpainting. White regions are regenerated, black regions are preserved. Requires image_url.
15463
15246
  */
15464
- image_size?:
15465
- | Components.ImageSize
15466
- | (
15467
- | 'square_hd'
15468
- | 'square'
15469
- | 'portrait_4_3'
15470
- | 'portrait_16_9'
15471
- | 'landscape_4_3'
15472
- | 'landscape_16_9'
15473
- );
15247
+ mask_image_url?: string;
15474
15248
  /**
15475
15249
  * Number of Images
15476
15250
  * @description The number of images to generate.
@@ -15502,6 +15276,12 @@ export interface ZImageTurboTilingInput {
15502
15276
  * will output the same image every time.
15503
15277
  */
15504
15278
  seed?: number;
15279
+ /**
15280
+ * Strength
15281
+ * @description How much to transform the input image. Only used when image_url is provided.
15282
+ * @default 0.6
15283
+ */
15284
+ strength?: number;
15505
15285
  /**
15506
15286
  * Sync Mode
15507
15287
  * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
@@ -15529,9 +15309,9 @@ export interface ZImageTurboTilingInput {
15529
15309
  tiling_mode?: 'both' | 'horizontal' | 'vertical';
15530
15310
  }
15531
15311
 
15532
- export interface ZImageTurboTilingOutput extends SharedType_390 {}
15312
+ export interface ZImageTurboTilingLoraOutput extends SharedType_390 {}
15533
15313
 
15534
- export interface ZImageTurboLoraInput {
15314
+ export interface ZImageTurboTilingInput {
15535
15315
  /**
15536
15316
  * Acceleration
15537
15317
  * @description The acceleration level to use.
@@ -15553,8 +15333,8 @@ export interface ZImageTurboLoraInput {
15553
15333
  enable_safety_checker?: boolean;
15554
15334
  /**
15555
15335
  * Image Size
15556
- * @description The size of the generated image.
15557
- * @default landscape_4_3
15336
+ * @description The size of the generated image. Use 'auto' to match the input image size (or 1024x1024 if no image).
15337
+ * @default square_hd
15558
15338
  */
15559
15339
  image_size?:
15560
15340
  | Components.ImageSize
@@ -15565,13 +15345,18 @@ export interface ZImageTurboLoraInput {
15565
15345
  | 'portrait_16_9'
15566
15346
  | 'landscape_4_3'
15567
15347
  | 'landscape_16_9'
15348
+ | 'auto'
15568
15349
  );
15569
15350
  /**
15570
- * Loras
15571
- * @description List of LoRA weights to apply (maximum 3).
15572
- * @default []
15351
+ * Image URL
15352
+ * @description URL of an image for image-to-image or inpainting. When provided without mask_image_url, performs image-to-image; with mask_image_url, performs inpainting.
15573
15353
  */
15574
- loras?: Components.LoRAInput_2[];
15354
+ image_url?: string;
15355
+ /**
15356
+ * Mask Image URL
15357
+ * @description URL of a mask image for inpainting. White regions are regenerated, black regions are preserved. Requires image_url.
15358
+ */
15359
+ mask_image_url?: string;
15575
15360
  /**
15576
15361
  * Number of Images
15577
15362
  * @description The number of images to generate.
@@ -15594,7 +15379,7 @@ export interface ZImageTurboLoraInput {
15594
15379
  /**
15595
15380
  * Prompt
15596
15381
  * @description The prompt to generate an image from.
15597
- * @example A hyper-realistic, close-up portrait of a tribal elder from the Omo Valley, painted with intricate white chalk patterns and adorned with a headdress made of dried flowers, seed pods, and rusted bottle caps. The focus is razor-sharp on the texture of the skin, showing every pore, wrinkle, and scar that tells a story of survival. The background is a blurred, smoky hut interior, with the warm glow of a cooking fire reflecting in the subject's dark, soulful eyes. Shot on a Leica M6 with Kodak Portra 400 film grain aesthetic.
15382
+ * @example A hyper-realistic, high-resolution 4k texture of an ancient weathered brick wall heavily overgrown with lush green moss and soft lichens. The bricks are aged, featuring deep earthy tones, natural cracks, and gritty textures. Vibrant emerald moss fills the mortar lines and spills over the rough surfaces of the stones. Uniform, flat cinematic lighting ensures no harsh shadows, highlighting the intricate organic details and damp stone surfaces. The composition is a perfectly balanced overhead view, showcasing a rich tapestry of botanical growth and masonry craftsmanship with professional clarity and hyper-detailed grit.
15598
15383
  */
15599
15384
  prompt: string;
15600
15385
  /**
@@ -15603,17 +15388,42 @@ export interface ZImageTurboLoraInput {
15603
15388
  * will output the same image every time.
15604
15389
  */
15605
15390
  seed?: number;
15391
+ /**
15392
+ * Strength
15393
+ * @description How much to transform the input image. Only used when image_url is provided.
15394
+ * @default 0.6
15395
+ */
15396
+ strength?: number;
15606
15397
  /**
15607
15398
  * Sync Mode
15608
15399
  * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
15609
15400
  * @default false
15610
15401
  */
15611
15402
  sync_mode?: boolean;
15403
+ /**
15404
+ * Tile Size
15405
+ * @description Tile size in latent space (64 = 512px, 128 = 1024px, 256 = 2048px).
15406
+ * @default 128
15407
+ */
15408
+ tile_size?: number;
15409
+ /**
15410
+ * Tile Stride
15411
+ * @description Tile stride in latent space. (32 = 256px, 64 = 512px, 128 = 1024px).
15412
+ * @default 64
15413
+ */
15414
+ tile_stride?: number;
15415
+ /**
15416
+ * Tiling Mode
15417
+ * @description Tiling direction: 'both' (omnidirectional), 'horizontal' (left-right only), 'vertical' (top-bottom only).
15418
+ * @default both
15419
+ * @enum {string}
15420
+ */
15421
+ tiling_mode?: 'both' | 'horizontal' | 'vertical';
15612
15422
  }
15613
15423
 
15614
- export interface ZImageTurboLoraOutput extends SharedType_352 {}
15424
+ export interface ZImageTurboTilingOutput extends SharedType_390 {}
15615
15425
 
15616
- export interface ZImageTurboInpaintLoraInput {
15426
+ export interface ZImageTurboLoraInput {
15617
15427
  /**
15618
15428
  * Acceleration
15619
15429
  * @description The acceleration level to use.
@@ -15621,24 +15431,6 @@ export interface ZImageTurboInpaintLoraInput {
15621
15431
  * @enum {string}
15622
15432
  */
15623
15433
  acceleration?: 'none' | 'regular' | 'high';
15624
- /**
15625
- * Control End
15626
- * @description The end of the controlnet conditioning.
15627
- * @default 0.8
15628
- */
15629
- control_end?: number;
15630
- /**
15631
- * Control Scale
15632
- * @description The scale of the controlnet conditioning.
15633
- * @default 0.75
15634
- */
15635
- control_scale?: number;
15636
- /**
15637
- * Control Start
15638
- * @description The start of the controlnet conditioning.
15639
- * @default 0
15640
- */
15641
- control_start?: number;
15642
15434
  /**
15643
15435
  * Enable Prompt Expansion
15644
15436
  * @description Whether to enable prompt expansion. Note: this will increase the price by 0.0025 credits per request.
@@ -15654,7 +15446,7 @@ export interface ZImageTurboInpaintLoraInput {
15654
15446
  /**
15655
15447
  * Image Size
15656
15448
  * @description The size of the generated image.
15657
- * @default auto
15449
+ * @default landscape_4_3
15658
15450
  */
15659
15451
  image_size?:
15660
15452
  | Components.ImageSize
@@ -15665,26 +15457,13 @@ export interface ZImageTurboInpaintLoraInput {
15665
15457
  | 'portrait_16_9'
15666
15458
  | 'landscape_4_3'
15667
15459
  | 'landscape_16_9'
15668
- | 'auto'
15669
15460
  );
15670
- /**
15671
- * Image URL
15672
- * @description URL of Image for Inpaint generation.
15673
- * @example https://storage.googleapis.com/falserverless/example_inputs/inpaint-input.jpg
15674
- */
15675
- image_url: string;
15676
15461
  /**
15677
15462
  * Loras
15678
15463
  * @description List of LoRA weights to apply (maximum 3).
15679
15464
  * @default []
15680
15465
  */
15681
15466
  loras?: Components.LoRAInput_2[];
15682
- /**
15683
- * Mask Image URL
15684
- * @description URL of Mask for Inpaint generation.
15685
- * @example https://storage.googleapis.com/falserverless/whls/z-image-inpaint-mask.jpg
15686
- */
15687
- mask_image_url: string;
15688
15467
  /**
15689
15468
  * Number of Images
15690
15469
  * @description The number of images to generate.
@@ -15707,7 +15486,7 @@ export interface ZImageTurboInpaintLoraInput {
15707
15486
  /**
15708
15487
  * Prompt
15709
15488
  * @description The prompt to generate an image from.
15710
- * @example A young Asian woman with long, vibrant purple hair stands on a sunlit sandy beach, posing confidently with her left hand resting on her hip. She gazes directly at the camera with a neutral expression. A sleek black ribbon bow is tied neatly on the right side of her head, just above her ear. She wears a flowing white cotton dress with a fitted bodice and a flared skirt that reaches mid-calf, slightly lifted by a gentle sea breeze. The beach behind her features fine, pale golden sand with subtle footprints, leading to calm turquoise waves under a clear blue sky with soft, wispy clouds. The lighting is natural daylight, casting soft shadows to her left, indicating late afternoon sun. The horizon line is visible in the background, with a faint silhouette of distant dunes. Her skin tone is fair with a natural glow, and her facial features are delicately defined. The composition is centered on her figure, framed from mid-thigh up, with shallow depth of field blurring the distant waves slightly.
15489
+ * @example A hyper-realistic, close-up portrait of a tribal elder from the Omo Valley, painted with intricate white chalk patterns and adorned with a headdress made of dried flowers, seed pods, and rusted bottle caps. The focus is razor-sharp on the texture of the skin, showing every pore, wrinkle, and scar that tells a story of survival. The background is a blurred, smoky hut interior, with the warm glow of a cooking fire reflecting in the subject's dark, soulful eyes. Shot on a Leica M6 with Kodak Portra 400 film grain aesthetic.
15711
15490
  */
15712
15491
  prompt: string;
15713
15492
  /**
@@ -15716,12 +15495,6 @@ export interface ZImageTurboInpaintLoraInput {
15716
15495
  * will output the same image every time.
15717
15496
  */
15718
15497
  seed?: number;
15719
- /**
15720
- * Strength
15721
- * @description The strength of the inpaint conditioning.
15722
- * @default 1
15723
- */
15724
- strength?: number;
15725
15498
  /**
15726
15499
  * Sync Mode
15727
15500
  * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
@@ -15730,9 +15503,9 @@ export interface ZImageTurboInpaintLoraInput {
15730
15503
  sync_mode?: boolean;
15731
15504
  }
15732
15505
 
15733
- export interface ZImageTurboInpaintLoraOutput extends SharedType_18d {}
15506
+ export interface ZImageTurboLoraOutput extends SharedType_352 {}
15734
15507
 
15735
- export interface ZImageTurboInpaintInput {
15508
+ export interface ZImageTurboInpaintLoraInput {
15736
15509
  /**
15737
15510
  * Acceleration
15738
15511
  * @description The acceleration level to use.
@@ -15792,6 +15565,12 @@ export interface ZImageTurboInpaintInput {
15792
15565
  * @example https://storage.googleapis.com/falserverless/example_inputs/inpaint-input.jpg
15793
15566
  */
15794
15567
  image_url: string;
15568
+ /**
15569
+ * Loras
15570
+ * @description List of LoRA weights to apply (maximum 3).
15571
+ * @default []
15572
+ */
15573
+ loras?: Components.LoRAInput_2[];
15795
15574
  /**
15796
15575
  * Mask Image URL
15797
15576
  * @description URL of Mask for Inpaint generation.
@@ -15843,9 +15622,9 @@ export interface ZImageTurboInpaintInput {
15843
15622
  sync_mode?: boolean;
15844
15623
  }
15845
15624
 
15846
- export interface ZImageTurboInpaintOutput extends SharedType_18d {}
15625
+ export interface ZImageTurboInpaintLoraOutput extends SharedType_18d {}
15847
15626
 
15848
- export interface ZImageTurboImageToImageLoraInput {
15627
+ export interface ZImageTurboInpaintInput {
15849
15628
  /**
15850
15629
  * Acceleration
15851
15630
  * @description The acceleration level to use.
@@ -15854,100 +15633,23 @@ export interface ZImageTurboImageToImageLoraInput {
15854
15633
  */
15855
15634
  acceleration?: 'none' | 'regular' | 'high';
15856
15635
  /**
15857
- * Enable Prompt Expansion
15858
- * @description Whether to enable prompt expansion. Note: this will increase the price by 0.0025 credits per request.
15859
- * @default false
15860
- */
15861
- enable_prompt_expansion?: boolean;
15862
- /**
15863
- * Enable Safety Checker
15864
- * @description If set to true, the safety checker will be enabled.
15865
- * @default true
15866
- */
15867
- enable_safety_checker?: boolean;
15868
- /**
15869
- * Image Size
15870
- * @description The size of the generated image.
15871
- * @default auto
15872
- */
15873
- image_size?:
15874
- | Components.ImageSize
15875
- | (
15876
- | 'square_hd'
15877
- | 'square'
15878
- | 'portrait_4_3'
15879
- | 'portrait_16_9'
15880
- | 'landscape_4_3'
15881
- | 'landscape_16_9'
15882
- | 'auto'
15883
- );
15884
- /**
15885
- * Image URL
15886
- * @description URL of Image for Image-to-Image generation.
15887
- * @example https://storage.googleapis.com/falserverless/example_inputs/z-image-turbo-i2i-input.png
15888
- */
15889
- image_url: string;
15890
- /**
15891
- * Loras
15892
- * @description List of LoRA weights to apply (maximum 3).
15893
- * @default []
15894
- */
15895
- loras?: Components.LoRAInput_2[];
15896
- /**
15897
- * Number of Images
15898
- * @description The number of images to generate.
15899
- * @default 1
15900
- */
15901
- num_images?: number;
15902
- /**
15903
- * Number of Inference Steps
15904
- * @description The number of inference steps to perform.
15905
- * @default 8
15906
- */
15907
- num_inference_steps?: number;
15908
- /**
15909
- * Output Format
15910
- * @description The format of the generated image.
15911
- * @default png
15912
- * @enum {string}
15913
- */
15914
- output_format?: 'jpeg' | 'png' | 'webp';
15915
- /**
15916
- * Prompt
15917
- * @description The prompt to generate an image from.
15918
- * @example A young Asian woman with long, vibrant purple hair stands on a sunlit sandy beach, posing confidently with her left hand resting on her hip. She gazes directly at the camera with a neutral expression. A sleek black ribbon bow is tied neatly on the right side of her head, just above her ear. She wears a flowing white cotton dress with a fitted bodice and a flared skirt that reaches mid-calf, slightly lifted by a gentle sea breeze. The beach behind her features fine, pale golden sand with subtle footprints, leading to calm turquoise waves under a clear blue sky with soft, wispy clouds. The lighting is natural daylight, casting soft shadows to her left, indicating late afternoon sun. The horizon line is visible in the background, with a faint silhouette of distant dunes. Her skin tone is fair with a natural glow, and her facial features are delicately defined. The composition is centered on her figure, framed from mid-thigh up, with shallow depth of field blurring the distant waves slightly.
15919
- */
15920
- prompt: string;
15921
- /**
15922
- * Seed
15923
- * @description The same seed and the same prompt given to the same version of the model
15924
- * will output the same image every time.
15925
- */
15926
- seed?: number;
15927
- /**
15928
- * Strength
15929
- * @description The strength of the image-to-image conditioning.
15930
- * @default 0.6
15636
+ * Control End
15637
+ * @description The end of the controlnet conditioning.
15638
+ * @default 0.8
15931
15639
  */
15932
- strength?: number;
15640
+ control_end?: number;
15933
15641
  /**
15934
- * Sync Mode
15935
- * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
15936
- * @default false
15642
+ * Control Scale
15643
+ * @description The scale of the controlnet conditioning.
15644
+ * @default 0.75
15937
15645
  */
15938
- sync_mode?: boolean;
15939
- }
15940
-
15941
- export interface ZImageTurboImageToImageLoraOutput extends SharedType_6b3 {}
15942
-
15943
- export interface ZImageTurboImageToImageInput {
15646
+ control_scale?: number;
15944
15647
  /**
15945
- * Acceleration
15946
- * @description The acceleration level to use.
15947
- * @default regular
15948
- * @enum {string}
15648
+ * Control Start
15649
+ * @description The start of the controlnet conditioning.
15650
+ * @default 0
15949
15651
  */
15950
- acceleration?: 'none' | 'regular' | 'high';
15652
+ control_start?: number;
15951
15653
  /**
15952
15654
  * Enable Prompt Expansion
15953
15655
  * @description Whether to enable prompt expansion. Note: this will increase the price by 0.0025 credits per request.
@@ -15978,10 +15680,16 @@ export interface ZImageTurboImageToImageInput {
15978
15680
  );
15979
15681
  /**
15980
15682
  * Image URL
15981
- * @description URL of Image for Image-to-Image generation.
15982
- * @example https://storage.googleapis.com/falserverless/example_inputs/z-image-turbo-i2i-input.png
15683
+ * @description URL of Image for Inpaint generation.
15684
+ * @example https://storage.googleapis.com/falserverless/example_inputs/inpaint-input.jpg
15983
15685
  */
15984
15686
  image_url: string;
15687
+ /**
15688
+ * Mask Image URL
15689
+ * @description URL of Mask for Inpaint generation.
15690
+ * @example https://storage.googleapis.com/falserverless/whls/z-image-inpaint-mask.jpg
15691
+ */
15692
+ mask_image_url: string;
15985
15693
  /**
15986
15694
  * Number of Images
15987
15695
  * @description The number of images to generate.
@@ -16015,8 +15723,8 @@ export interface ZImageTurboImageToImageInput {
16015
15723
  seed?: number;
16016
15724
  /**
16017
15725
  * Strength
16018
- * @description The strength of the image-to-image conditioning.
16019
- * @default 0.6
15726
+ * @description The strength of the inpaint conditioning.
15727
+ * @default 1
16020
15728
  */
16021
15729
  strength?: number;
16022
15730
  /**
@@ -16027,9 +15735,9 @@ export interface ZImageTurboImageToImageInput {
16027
15735
  sync_mode?: boolean;
16028
15736
  }
16029
15737
 
16030
- export interface ZImageTurboImageToImageOutput extends SharedType_6b3 {}
15738
+ export interface ZImageTurboInpaintOutput extends SharedType_18d {}
16031
15739
 
16032
- export interface ZImageTurboControlnetLoraInput {
15740
+ export interface ZImageTurboImageToImageLoraInput {
16033
15741
  /**
16034
15742
  * Acceleration
16035
15743
  * @description The acceleration level to use.
@@ -16037,24 +15745,6 @@ export interface ZImageTurboControlnetLoraInput {
16037
15745
  * @enum {string}
16038
15746
  */
16039
15747
  acceleration?: 'none' | 'regular' | 'high';
16040
- /**
16041
- * Control End
16042
- * @description The end of the controlnet conditioning.
16043
- * @default 0.8
16044
- */
16045
- control_end?: number;
16046
- /**
16047
- * Control Scale
16048
- * @description The scale of the controlnet conditioning.
16049
- * @default 0.75
16050
- */
16051
- control_scale?: number;
16052
- /**
16053
- * Control Start
16054
- * @description The start of the controlnet conditioning.
16055
- * @default 0
16056
- */
16057
- control_start?: number;
16058
15748
  /**
16059
15749
  * Enable Prompt Expansion
16060
15750
  * @description Whether to enable prompt expansion. Note: this will increase the price by 0.0025 credits per request.
@@ -16085,8 +15775,8 @@ export interface ZImageTurboControlnetLoraInput {
16085
15775
  );
16086
15776
  /**
16087
15777
  * Image URL
16088
- * @description URL of Image for ControlNet generation.
16089
- * @example https://storage.googleapis.com/falserverless/example_inputs/z-image-turbo-controlnet-input.jpg
15778
+ * @description URL of Image for Image-to-Image generation.
15779
+ * @example https://storage.googleapis.com/falserverless/example_inputs/z-image-turbo-i2i-input.png
16090
15780
  */
16091
15781
  image_url: string;
16092
15782
  /**
@@ -16114,17 +15804,10 @@ export interface ZImageTurboControlnetLoraInput {
16114
15804
  * @enum {string}
16115
15805
  */
16116
15806
  output_format?: 'jpeg' | 'png' | 'webp';
16117
- /**
16118
- * Preprocess
16119
- * @description What kind of preprocessing to apply to the image, if any.
16120
- * @default none
16121
- * @example none
16122
- */
16123
- preprocess?: 'none' | 'canny' | 'depth' | 'pose';
16124
15807
  /**
16125
15808
  * Prompt
16126
15809
  * @description The prompt to generate an image from.
16127
- * @example A single leopard, its spotted golden coat detailed with black rosettes, cautiously peeks its head through dense green foliage. The leopard’s eyes are alert and focused forward, ears perked, whiskers slightly visible. The bushes consist of thick, leafy shrubs with varying shades of green, some leaves partially obscuring the leopard’s muzzle and forehead. Soft natural daylight filters through the canopy above, casting dappled shadows across the animal’s fur and surrounding leaves. The composition is a medium close-up, centered on the leopard’s head emerging from the undergrowth, with shallow depth of field blurring the background vegetation.
15810
+ * @example A young Asian woman with long, vibrant purple hair stands on a sunlit sandy beach, posing confidently with her left hand resting on her hip. She gazes directly at the camera with a neutral expression. A sleek black ribbon bow is tied neatly on the right side of her head, just above her ear. She wears a flowing white cotton dress with a fitted bodice and a flared skirt that reaches mid-calf, slightly lifted by a gentle sea breeze. The beach behind her features fine, pale golden sand with subtle footprints, leading to calm turquoise waves under a clear blue sky with soft, wispy clouds. The lighting is natural daylight, casting soft shadows to her left, indicating late afternoon sun. The horizon line is visible in the background, with a faint silhouette of distant dunes. Her skin tone is fair with a natural glow, and her facial features are delicately defined. The composition is centered on her figure, framed from mid-thigh up, with shallow depth of field blurring the distant waves slightly.
16128
15811
  */
16129
15812
  prompt: string;
16130
15813
  /**
@@ -16133,6 +15816,12 @@ export interface ZImageTurboControlnetLoraInput {
16133
15816
  * will output the same image every time.
16134
15817
  */
16135
15818
  seed?: number;
15819
+ /**
15820
+ * Strength
15821
+ * @description The strength of the image-to-image conditioning.
15822
+ * @default 0.6
15823
+ */
15824
+ strength?: number;
16136
15825
  /**
16137
15826
  * Sync Mode
16138
15827
  * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
@@ -16141,9 +15830,9 @@ export interface ZImageTurboControlnetLoraInput {
16141
15830
  sync_mode?: boolean;
16142
15831
  }
16143
15832
 
16144
- export interface ZImageTurboControlnetLoraOutput extends SharedType_576 {}
15833
+ export interface ZImageTurboImageToImageLoraOutput extends SharedType_6b3 {}
16145
15834
 
16146
- export interface ZImageTurboControlnetInput {
15835
+ export interface ZImageTurboImageToImageInput {
16147
15836
  /**
16148
15837
  * Acceleration
16149
15838
  * @description The acceleration level to use.
@@ -16151,24 +15840,6 @@ export interface ZImageTurboControlnetInput {
16151
15840
  * @enum {string}
16152
15841
  */
16153
15842
  acceleration?: 'none' | 'regular' | 'high';
16154
- /**
16155
- * Control End
16156
- * @description The end of the controlnet conditioning.
16157
- * @default 0.8
16158
- */
16159
- control_end?: number;
16160
- /**
16161
- * Control Scale
16162
- * @description The scale of the controlnet conditioning.
16163
- * @default 0.75
16164
- */
16165
- control_scale?: number;
16166
- /**
16167
- * Control Start
16168
- * @description The start of the controlnet conditioning.
16169
- * @default 0
16170
- */
16171
- control_start?: number;
16172
15843
  /**
16173
15844
  * Enable Prompt Expansion
16174
15845
  * @description Whether to enable prompt expansion. Note: this will increase the price by 0.0025 credits per request.
@@ -16199,8 +15870,229 @@ export interface ZImageTurboControlnetInput {
16199
15870
  );
16200
15871
  /**
16201
15872
  * Image URL
16202
- * @description URL of Image for ControlNet generation.
16203
- * @example https://storage.googleapis.com/falserverless/example_inputs/z-image-turbo-controlnet-input.jpg
15873
+ * @description URL of Image for Image-to-Image generation.
15874
+ * @example https://storage.googleapis.com/falserverless/example_inputs/z-image-turbo-i2i-input.png
15875
+ */
15876
+ image_url: string;
15877
+ /**
15878
+ * Number of Images
15879
+ * @description The number of images to generate.
15880
+ * @default 1
15881
+ */
15882
+ num_images?: number;
15883
+ /**
15884
+ * Number of Inference Steps
15885
+ * @description The number of inference steps to perform.
15886
+ * @default 8
15887
+ */
15888
+ num_inference_steps?: number;
15889
+ /**
15890
+ * Output Format
15891
+ * @description The format of the generated image.
15892
+ * @default png
15893
+ * @enum {string}
15894
+ */
15895
+ output_format?: 'jpeg' | 'png' | 'webp';
15896
+ /**
15897
+ * Prompt
15898
+ * @description The prompt to generate an image from.
15899
+ * @example A young Asian woman with long, vibrant purple hair stands on a sunlit sandy beach, posing confidently with her left hand resting on her hip. She gazes directly at the camera with a neutral expression. A sleek black ribbon bow is tied neatly on the right side of her head, just above her ear. She wears a flowing white cotton dress with a fitted bodice and a flared skirt that reaches mid-calf, slightly lifted by a gentle sea breeze. The beach behind her features fine, pale golden sand with subtle footprints, leading to calm turquoise waves under a clear blue sky with soft, wispy clouds. The lighting is natural daylight, casting soft shadows to her left, indicating late afternoon sun. The horizon line is visible in the background, with a faint silhouette of distant dunes. Her skin tone is fair with a natural glow, and her facial features are delicately defined. The composition is centered on her figure, framed from mid-thigh up, with shallow depth of field blurring the distant waves slightly.
15900
+ */
15901
+ prompt: string;
15902
+ /**
15903
+ * Seed
15904
+ * @description The same seed and the same prompt given to the same version of the model
15905
+ * will output the same image every time.
15906
+ */
15907
+ seed?: number;
15908
+ /**
15909
+ * Strength
15910
+ * @description The strength of the image-to-image conditioning.
15911
+ * @default 0.6
15912
+ */
15913
+ strength?: number;
15914
+ /**
15915
+ * Sync Mode
15916
+ * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
15917
+ * @default false
15918
+ */
15919
+ sync_mode?: boolean;
15920
+ }
15921
+
15922
+ export interface ZImageTurboImageToImageOutput extends SharedType_6b3 {}
15923
+
15924
+ export interface ZImageTurboControlnetLoraInput {
15925
+ /**
15926
+ * Acceleration
15927
+ * @description The acceleration level to use.
15928
+ * @default regular
15929
+ * @enum {string}
15930
+ */
15931
+ acceleration?: 'none' | 'regular' | 'high';
15932
+ /**
15933
+ * Control End
15934
+ * @description The end of the controlnet conditioning.
15935
+ * @default 0.8
15936
+ */
15937
+ control_end?: number;
15938
+ /**
15939
+ * Control Scale
15940
+ * @description The scale of the controlnet conditioning.
15941
+ * @default 0.75
15942
+ */
15943
+ control_scale?: number;
15944
+ /**
15945
+ * Control Start
15946
+ * @description The start of the controlnet conditioning.
15947
+ * @default 0
15948
+ */
15949
+ control_start?: number;
15950
+ /**
15951
+ * Enable Prompt Expansion
15952
+ * @description Whether to enable prompt expansion. Note: this will increase the price by 0.0025 credits per request.
15953
+ * @default false
15954
+ */
15955
+ enable_prompt_expansion?: boolean;
15956
+ /**
15957
+ * Enable Safety Checker
15958
+ * @description If set to true, the safety checker will be enabled.
15959
+ * @default true
15960
+ */
15961
+ enable_safety_checker?: boolean;
15962
+ /**
15963
+ * Image Size
15964
+ * @description The size of the generated image.
15965
+ * @default auto
15966
+ */
15967
+ image_size?:
15968
+ | Components.ImageSize
15969
+ | (
15970
+ | 'square_hd'
15971
+ | 'square'
15972
+ | 'portrait_4_3'
15973
+ | 'portrait_16_9'
15974
+ | 'landscape_4_3'
15975
+ | 'landscape_16_9'
15976
+ | 'auto'
15977
+ );
15978
+ /**
15979
+ * Image URL
15980
+ * @description URL of Image for ControlNet generation.
15981
+ * @example https://storage.googleapis.com/falserverless/example_inputs/z-image-turbo-controlnet-input.jpg
15982
+ */
15983
+ image_url: string;
15984
+ /**
15985
+ * Loras
15986
+ * @description List of LoRA weights to apply (maximum 3).
15987
+ * @default []
15988
+ */
15989
+ loras?: Components.LoRAInput_2[];
15990
+ /**
15991
+ * Number of Images
15992
+ * @description The number of images to generate.
15993
+ * @default 1
15994
+ */
15995
+ num_images?: number;
15996
+ /**
15997
+ * Number of Inference Steps
15998
+ * @description The number of inference steps to perform.
15999
+ * @default 8
16000
+ */
16001
+ num_inference_steps?: number;
16002
+ /**
16003
+ * Output Format
16004
+ * @description The format of the generated image.
16005
+ * @default png
16006
+ * @enum {string}
16007
+ */
16008
+ output_format?: 'jpeg' | 'png' | 'webp';
16009
+ /**
16010
+ * Preprocess
16011
+ * @description What kind of preprocessing to apply to the image, if any.
16012
+ * @default none
16013
+ * @example none
16014
+ */
16015
+ preprocess?: 'none' | 'canny' | 'depth' | 'pose';
16016
+ /**
16017
+ * Prompt
16018
+ * @description The prompt to generate an image from.
16019
+ * @example A single leopard, its spotted golden coat detailed with black rosettes, cautiously peeks its head through dense green foliage. The leopard’s eyes are alert and focused forward, ears perked, whiskers slightly visible. The bushes consist of thick, leafy shrubs with varying shades of green, some leaves partially obscuring the leopard’s muzzle and forehead. Soft natural daylight filters through the canopy above, casting dappled shadows across the animal’s fur and surrounding leaves. The composition is a medium close-up, centered on the leopard’s head emerging from the undergrowth, with shallow depth of field blurring the background vegetation.
16020
+ */
16021
+ prompt: string;
16022
+ /**
16023
+ * Seed
16024
+ * @description The same seed and the same prompt given to the same version of the model
16025
+ * will output the same image every time.
16026
+ */
16027
+ seed?: number;
16028
+ /**
16029
+ * Sync Mode
16030
+ * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
16031
+ * @default false
16032
+ */
16033
+ sync_mode?: boolean;
16034
+ }
16035
+
16036
+ export interface ZImageTurboControlnetLoraOutput extends SharedType_576 {}
16037
+
16038
+ export interface ZImageTurboControlnetInput {
16039
+ /**
16040
+ * Acceleration
16041
+ * @description The acceleration level to use.
16042
+ * @default regular
16043
+ * @enum {string}
16044
+ */
16045
+ acceleration?: 'none' | 'regular' | 'high';
16046
+ /**
16047
+ * Control End
16048
+ * @description The end of the controlnet conditioning.
16049
+ * @default 0.8
16050
+ */
16051
+ control_end?: number;
16052
+ /**
16053
+ * Control Scale
16054
+ * @description The scale of the controlnet conditioning.
16055
+ * @default 0.75
16056
+ */
16057
+ control_scale?: number;
16058
+ /**
16059
+ * Control Start
16060
+ * @description The start of the controlnet conditioning.
16061
+ * @default 0
16062
+ */
16063
+ control_start?: number;
16064
+ /**
16065
+ * Enable Prompt Expansion
16066
+ * @description Whether to enable prompt expansion. Note: this will increase the price by 0.0025 credits per request.
16067
+ * @default false
16068
+ */
16069
+ enable_prompt_expansion?: boolean;
16070
+ /**
16071
+ * Enable Safety Checker
16072
+ * @description If set to true, the safety checker will be enabled.
16073
+ * @default true
16074
+ */
16075
+ enable_safety_checker?: boolean;
16076
+ /**
16077
+ * Image Size
16078
+ * @description The size of the generated image.
16079
+ * @default auto
16080
+ */
16081
+ image_size?:
16082
+ | Components.ImageSize
16083
+ | (
16084
+ | 'square_hd'
16085
+ | 'square'
16086
+ | 'portrait_4_3'
16087
+ | 'portrait_16_9'
16088
+ | 'landscape_4_3'
16089
+ | 'landscape_16_9'
16090
+ | 'auto'
16091
+ );
16092
+ /**
16093
+ * Image URL
16094
+ * @description URL of Image for ControlNet generation.
16095
+ * @example https://storage.googleapis.com/falserverless/example_inputs/z-image-turbo-controlnet-input.jpg
16204
16096
  */
16205
16097
  image_url: string;
16206
16098
  /**
@@ -25483,7 +25375,6 @@ export interface SteadyDancerOutput {
25483
25375
  */
25484
25376
  seed: number;
25485
25377
  /**
25486
- * Video
25487
25378
  * @description The generated dance animation video.
25488
25379
  * @example {
25489
25380
  * "file_size": 7772111,
@@ -25492,7 +25383,7 @@ export interface SteadyDancerOutput {
25492
25383
  * "url": "https://v3b.fal.media/files/b/0a87871b/ll5ps0ZyBgxBkuWz-fHcT_output_with_audio.mp4"
25493
25384
  * }
25494
25385
  */
25495
- video: Components.File_1;
25386
+ video: Components.File;
25496
25387
  }
25497
25388
 
25498
25389
  export interface StarVectorInput {
@@ -26009,6 +25900,7 @@ export interface StableDiffusionV15Input {
26009
25900
  * Seed
26010
25901
  * @description The same seed and the same prompt given to the same version of Stable Diffusion
26011
25902
  * will output the same image every time.
25903
+ * @default null
26012
25904
  */
26013
25905
  seed?: number;
26014
25906
  /**
@@ -28614,6 +28506,13 @@ export interface Sam3VideoInput {
28614
28506
  * @description [DEPRECATED] Use 'prompt' instead. Kept for backward compatibility.
28615
28507
  */
28616
28508
  text_prompt?: string;
28509
+ /**
28510
+ * Video Output Type
28511
+ * @description The output type of the generated video.
28512
+ * @default X264 (.mp4)
28513
+ * @enum {string}
28514
+ */
28515
+ video_output_type?: 'X264 (.mp4)' | 'VP9 (.webm)';
28617
28516
  /**
28618
28517
  * Video Url
28619
28518
  * @description The URL of the video to be segmented.
@@ -36099,6 +35998,98 @@ export interface PhotomakerOutput {
36099
35998
  seed: number;
36100
35999
  }
36101
36000
 
36001
+ export interface PersonaplexRealtimeInput {
36002
+ /**
36003
+ * Audio
36004
+ * Format: binary
36005
+ * @description Input audio chunk (PCM s16le, 24kHz mono). Base64-encoded in JSON transport.
36006
+ */
36007
+ audio: string;
36008
+ /**
36009
+ * Prompt
36010
+ * @description Text prompt describing the AI persona and conversation context.
36011
+ * @default You are a wise and friendly teacher. Answer questions or provide advice in a clear and engaging way.
36012
+ * @example You work for SwiftPlex Appliances which is a appliance repair company and your name is Farhod Toshmatov. Information: The dishwasher model is out of stock for replacement parts; we can use an alternative part with a 3-day delay. Labor cost remains $60 per hour.
36013
+ */
36014
+ prompt?: string;
36015
+ /**
36016
+ * Seed
36017
+ * @description Random seed for reproducibility.
36018
+ * @default null
36019
+ */
36020
+ seed?: number;
36021
+ /**
36022
+ * Temperature Audio
36023
+ * @description Audio sampling temperature. Higher values produce more diverse outputs.
36024
+ * @default 0.8
36025
+ */
36026
+ temperature_audio?: number;
36027
+ /**
36028
+ * Temperature Text
36029
+ * @description Text sampling temperature. Higher values produce more diverse outputs.
36030
+ * @default 0.7
36031
+ */
36032
+ temperature_text?: number;
36033
+ /**
36034
+ * Top K Audio
36035
+ * @description Top-K sampling for audio tokens.
36036
+ * @default 250
36037
+ */
36038
+ top_k_audio?: number;
36039
+ /**
36040
+ * Top K Text
36041
+ * @description Top-K sampling for text tokens.
36042
+ * @default 25
36043
+ */
36044
+ top_k_text?: number;
36045
+ /**
36046
+ * Voice
36047
+ * @description Voice ID for the AI response. NAT = natural, VAR = variety. F = female, M = male. Ignored when voice_audio_url is provided.
36048
+ * @default NATF2
36049
+ * @enum {string}
36050
+ */
36051
+ voice?:
36052
+ | 'NATF0'
36053
+ | 'NATF1'
36054
+ | 'NATF2'
36055
+ | 'NATF3'
36056
+ | 'NATM0'
36057
+ | 'NATM1'
36058
+ | 'NATM2'
36059
+ | 'NATM3'
36060
+ | 'VARF0'
36061
+ | 'VARF1'
36062
+ | 'VARF2'
36063
+ | 'VARF3'
36064
+ | 'VARF4'
36065
+ | 'VARM0'
36066
+ | 'VARM1'
36067
+ | 'VARM2'
36068
+ | 'VARM3'
36069
+ | 'VARM4';
36070
+ /**
36071
+ * Voice Audio Url
36072
+ * @description URL to a voice sample audio for on-the-fly voice cloning. When provided, the AI responds in the cloned voice instead of the preset 'voice'. 10+ seconds of clear speech recommended. Billed at 2x rate.
36073
+ * @default null
36074
+ */
36075
+ voice_audio_url?: string;
36076
+ }
36077
+
36078
+ export interface PersonaplexRealtimeOutput {
36079
+ /**
36080
+ * Audio
36081
+ * Format: binary
36082
+ * @description Generated audio chunk (PCM s16le, 24kHz mono). Base64-encoded in JSON transport.
36083
+ */
36084
+ audio: string;
36085
+ /**
36086
+ * Text
36087
+ * @description Generated text tokens for this chunk.
36088
+ * @default
36089
+ */
36090
+ text?: string;
36091
+ }
36092
+
36102
36093
  export interface PersonaplexInput {
36103
36094
  /**
36104
36095
  * Audio Url
@@ -37284,13 +37275,13 @@ export interface NovaSrOutput {
37284
37275
  /**
37285
37276
  * @description The enhanced audio file.
37286
37277
  * @example {
37287
- * "channels": 1,
37278
+ * "bitrate": "192k",
37288
37279
  * "duration": 12.283291666666667,
37289
37280
  * "url": "https://v3b.fal.media/files/b/0a8a3f1a/lTKExJu-R6ZJdnFlpzEeq_TxmNTNhl.mp3",
37290
37281
  * "file_name": "lTKExJu-R6ZJdnFlpzEeq_TxmNTNhl.mp3",
37291
37282
  * "sample_rate": 48000,
37292
37283
  * "content_type": "audio/mpeg",
37293
- * "bitrate": "192k"
37284
+ * "channels": 1
37294
37285
  * }
37295
37286
  */
37296
37287
  audio: Components.AudioFile;
@@ -40751,7 +40742,147 @@ export interface Ltxv13b098DistilledImageToVideoInput extends SharedType_2e6 {}
40751
40742
 
40752
40743
  export interface Ltxv13b098DistilledImageToVideoOutput extends SharedType_6711 {}
40753
40744
 
40754
- export interface Ltxv13b098DistilledExtendInput extends SharedType_133 {}
40745
+ export interface Ltxv13b098DistilledExtendInput {
40746
+ /**
40747
+ * Aspect Ratio
40748
+ * @description The aspect ratio of the video.
40749
+ * @default auto
40750
+ * @example auto
40751
+ * @enum {string}
40752
+ */
40753
+ aspect_ratio?: '9:16' | '1:1' | '16:9' | 'auto';
40754
+ /**
40755
+ * Constant Rate Factor
40756
+ * @description The constant rate factor (CRF) to compress input media with. Compressed input media more closely matches the model's training data, which can improve motion quality.
40757
+ * @default 29
40758
+ * @example 29
40759
+ */
40760
+ constant_rate_factor?: number;
40761
+ /**
40762
+ * Enable Detail Pass
40763
+ * @description Whether to use a detail pass. If True, the model will perform a second pass to refine the video and enhance details. This incurs a 2.0x cost multiplier on the base price.
40764
+ * @default false
40765
+ * @example false
40766
+ */
40767
+ enable_detail_pass?: boolean;
40768
+ /**
40769
+ * Enable Safety Checker
40770
+ * @description Whether to enable the safety checker.
40771
+ * @default true
40772
+ * @example true
40773
+ */
40774
+ enable_safety_checker?: boolean;
40775
+ /**
40776
+ * Expand Prompt
40777
+ * @description Whether to expand the prompt using a language model.
40778
+ * @default false
40779
+ * @example false
40780
+ */
40781
+ expand_prompt?: boolean;
40782
+ /**
40783
+ * Number of Inference Steps
40784
+ * @description Number of inference steps during the first pass.
40785
+ * @default 8
40786
+ * @example 8
40787
+ */
40788
+ first_pass_num_inference_steps?: number;
40789
+ /**
40790
+ * Frame Rate
40791
+ * @description The frame rate of the video.
40792
+ * @default 24
40793
+ * @example 24
40794
+ */
40795
+ frame_rate?: number;
40796
+ /**
40797
+ * Loras
40798
+ * @description LoRA weights to use for generation
40799
+ * @default []
40800
+ */
40801
+ loras?: Components.LoRAWeight[];
40802
+ /**
40803
+ * Negative Prompt
40804
+ * @description Negative prompt for generation
40805
+ * @default worst quality, inconsistent motion, blurry, jittery, distorted
40806
+ */
40807
+ negative_prompt?: string;
40808
+ /**
40809
+ * Number of Frames
40810
+ * @description The number of frames in the video.
40811
+ * @default 121
40812
+ * @example 121
40813
+ */
40814
+ num_frames?: number;
40815
+ /**
40816
+ * Prompt
40817
+ * @description Text prompt to guide generation
40818
+ * @example Woman walking on a street in Tokyo
40819
+ */
40820
+ prompt: string;
40821
+ /**
40822
+ * Resolution
40823
+ * @description Resolution of the generated video.
40824
+ * @default 720p
40825
+ * @example 720p
40826
+ * @enum {string}
40827
+ */
40828
+ resolution?: '480p' | '720p';
40829
+ /**
40830
+ * Reverse Video
40831
+ * @description Whether to reverse the video.
40832
+ * @default false
40833
+ * @example false
40834
+ */
40835
+ reverse_video?: boolean;
40836
+ /**
40837
+ * Second Pass Number of Inference Steps
40838
+ * @description Number of inference steps during the second pass.
40839
+ * @default 8
40840
+ * @example 8
40841
+ */
40842
+ second_pass_num_inference_steps?: number;
40843
+ /**
40844
+ * Second Pass Skip Initial Steps
40845
+ * @description The number of inference steps to skip in the initial steps of the second pass. By skipping some steps at the beginning, the second pass can focus on smaller details instead of larger changes.
40846
+ * @default 5
40847
+ * @example 5
40848
+ */
40849
+ second_pass_skip_initial_steps?: number;
40850
+ /**
40851
+ * Seed
40852
+ * @description Random seed for generation
40853
+ */
40854
+ seed?: number;
40855
+ /**
40856
+ * Temporal AdaIN Factor
40857
+ * @description The factor for adaptive instance normalization (AdaIN) applied to generated video chunks after the first. This can help deal with a gradual increase in saturation/contrast in the generated video by normalizing the color distribution across the video. A high value will ensure the color distribution is more consistent across the video, while a low value will allow for more variation in color distribution.
40858
+ * @default 0.5
40859
+ * @example 0.5
40860
+ */
40861
+ temporal_adain_factor?: number;
40862
+ /**
40863
+ * Tone Map Compression Ratio
40864
+ * @description The compression ratio for tone mapping. This is used to compress the dynamic range of the video to improve visual quality. A value of 0.0 means no compression, while a value of 1.0 means maximum compression.
40865
+ * @default 0
40866
+ * @example 0
40867
+ */
40868
+ tone_map_compression_ratio?: number;
40869
+ /**
40870
+ * @description Video to be extended.
40871
+ * @example {
40872
+ * "video_url": "https://storage.googleapis.com/falserverless/web-examples/wan/t2v.mp4",
40873
+ * "start_frame_num": 0,
40874
+ * "reverse_video": false,
40875
+ * "limit_num_frames": false,
40876
+ * "resample_fps": false,
40877
+ * "strength": 1,
40878
+ * "target_fps": 24,
40879
+ * "max_num_frames": 1441,
40880
+ * "conditioning_type": "rgb",
40881
+ * "preprocess": false
40882
+ * }
40883
+ */
40884
+ video: Components.ExtendVideoConditioningInput;
40885
+ }
40755
40886
 
40756
40887
  export interface Ltxv13b098DistilledExtendOutput extends SharedType_844 {}
40757
40888
 
@@ -40787,6 +40918,12 @@ export interface Ltx2VideoTrainerInput {
40787
40918
  * @example false
40788
40919
  */
40789
40920
  auto_scale_input?: boolean;
40921
+ /**
40922
+ * Debug Dataset
40923
+ * @description When enabled, the trainer returns a downloadable archive of your preprocessed training data for manual inspection. Use this to verify that your videos, images, and captions were processed correctly before committing to a full training run.
40924
+ * @default false
40925
+ */
40926
+ debug_dataset?: boolean;
40790
40927
  /**
40791
40928
  * First Frame Conditioning P
40792
40929
  * @description Probability of conditioning on the first frame during training. Higher values improve image-to-video performance.
@@ -40949,6 +41086,12 @@ export interface Ltx2V2vTrainerInput {
40949
41086
  * @example false
40950
41087
  */
40951
41088
  auto_scale_input?: boolean;
41089
+ /**
41090
+ * Debug Dataset
41091
+ * @description When enabled, the trainer returns a downloadable archive of your preprocessed training data for manual inspection. Use this to verify that your videos, images, and captions were processed correctly before committing to a full training run.
41092
+ * @default false
41093
+ */
41094
+ debug_dataset?: boolean;
40952
41095
  /**
40953
41096
  * First Frame Conditioning P
40954
41097
  * @description Probability of conditioning on the first frame during training. Lower values work better for video-to-video transformation.
@@ -41729,7 +41872,147 @@ export interface LtxVideo13bDistilledImageToVideoInput extends SharedType_2e6 {}
41729
41872
 
41730
41873
  export interface LtxVideo13bDistilledImageToVideoOutput extends SharedType_6711 {}
41731
41874
 
41732
- export interface LtxVideo13bDistilledExtendInput extends SharedType_133 {}
41875
+ export interface LtxVideo13bDistilledExtendInput {
41876
+ /**
41877
+ * Aspect Ratio
41878
+ * @description The aspect ratio of the video.
41879
+ * @default auto
41880
+ * @example auto
41881
+ * @enum {string}
41882
+ */
41883
+ aspect_ratio?: '9:16' | '1:1' | '16:9' | 'auto';
41884
+ /**
41885
+ * Constant Rate Factor
41886
+ * @description The constant rate factor (CRF) to compress input media with. Compressed input media more closely matches the model's training data, which can improve motion quality.
41887
+ * @default 29
41888
+ * @example 29
41889
+ */
41890
+ constant_rate_factor?: number;
41891
+ /**
41892
+ * Enable Detail Pass
41893
+ * @description Whether to use a detail pass. If True, the model will perform a second pass to refine the video and enhance details. This incurs a 2.0x cost multiplier on the base price.
41894
+ * @default false
41895
+ * @example false
41896
+ */
41897
+ enable_detail_pass?: boolean;
41898
+ /**
41899
+ * Enable Safety Checker
41900
+ * @description Whether to enable the safety checker.
41901
+ * @default true
41902
+ * @example true
41903
+ */
41904
+ enable_safety_checker?: boolean;
41905
+ /**
41906
+ * Expand Prompt
41907
+ * @description Whether to expand the prompt using a language model.
41908
+ * @default false
41909
+ * @example false
41910
+ */
41911
+ expand_prompt?: boolean;
41912
+ /**
41913
+ * Number of Inference Steps
41914
+ * @description Number of inference steps during the first pass.
41915
+ * @default 8
41916
+ * @example 8
41917
+ */
41918
+ first_pass_num_inference_steps?: number;
41919
+ /**
41920
+ * Frame Rate
41921
+ * @description The frame rate of the video.
41922
+ * @default 24
41923
+ * @example 24
41924
+ */
41925
+ frame_rate?: number;
41926
+ /**
41927
+ * Loras
41928
+ * @description LoRA weights to use for generation
41929
+ * @default []
41930
+ */
41931
+ loras?: Components.LoRAWeight[];
41932
+ /**
41933
+ * Negative Prompt
41934
+ * @description Negative prompt for generation
41935
+ * @default worst quality, inconsistent motion, blurry, jittery, distorted
41936
+ */
41937
+ negative_prompt?: string;
41938
+ /**
41939
+ * Number of Frames
41940
+ * @description The number of frames in the video.
41941
+ * @default 121
41942
+ * @example 121
41943
+ */
41944
+ num_frames?: number;
41945
+ /**
41946
+ * Prompt
41947
+ * @description Text prompt to guide generation
41948
+ * @example Woman walking on a street in Tokyo
41949
+ */
41950
+ prompt: string;
41951
+ /**
41952
+ * Resolution
41953
+ * @description Resolution of the generated video.
41954
+ * @default 720p
41955
+ * @example 720p
41956
+ * @enum {string}
41957
+ */
41958
+ resolution?: '480p' | '720p';
41959
+ /**
41960
+ * Reverse Video
41961
+ * @description Whether to reverse the video.
41962
+ * @default false
41963
+ * @example false
41964
+ */
41965
+ reverse_video?: boolean;
41966
+ /**
41967
+ * Second Pass Number of Inference Steps
41968
+ * @description Number of inference steps during the second pass.
41969
+ * @default 8
41970
+ * @example 8
41971
+ */
41972
+ second_pass_num_inference_steps?: number;
41973
+ /**
41974
+ * Second Pass Skip Initial Steps
41975
+ * @description The number of inference steps to skip in the initial steps of the second pass. By skipping some steps at the beginning, the second pass can focus on smaller details instead of larger changes.
41976
+ * @default 5
41977
+ * @example 5
41978
+ */
41979
+ second_pass_skip_initial_steps?: number;
41980
+ /**
41981
+ * Seed
41982
+ * @description Random seed for generation
41983
+ */
41984
+ seed?: number;
41985
+ /**
41986
+ * Temporal AdaIN Factor
41987
+ * @description The factor for adaptive instance normalization (AdaIN) applied to generated video chunks after the first. This can help deal with a gradual increase in saturation/contrast in the generated video by normalizing the color distribution across the video. A high value will ensure the color distribution is more consistent across the video, while a low value will allow for more variation in color distribution.
41988
+ * @default 0.5
41989
+ * @example 0.5
41990
+ */
41991
+ temporal_adain_factor?: number;
41992
+ /**
41993
+ * Tone Map Compression Ratio
41994
+ * @description The compression ratio for tone mapping. This is used to compress the dynamic range of the video to improve visual quality. A value of 0.0 means no compression, while a value of 1.0 means maximum compression.
41995
+ * @default 0
41996
+ * @example 0
41997
+ */
41998
+ tone_map_compression_ratio?: number;
41999
+ /**
42000
+ * @description Video to be extended.
42001
+ * @example {
42002
+ * "video_url": "https://storage.googleapis.com/falserverless/web-examples/wan/t2v.mp4",
42003
+ * "reverse_video": false,
42004
+ * "start_frame_num": 0,
42005
+ * "limit_num_frames": false,
42006
+ * "resample_fps": false,
42007
+ * "strength": 1,
42008
+ * "target_fps": 24,
42009
+ * "max_num_frames": 1441,
42010
+ * "conditioning_type": "rgb",
42011
+ * "preprocess": false
42012
+ * }
42013
+ */
42014
+ video: Components.ExtendVideoConditioningInput;
42015
+ }
41733
42016
 
41734
42017
  export interface LtxVideo13bDistilledExtendOutput extends SharedType_844 {}
41735
42018
 
@@ -42148,8 +42431,8 @@ export interface LtxVideo13bDevExtendInput {
42148
42431
  * @description Video to be extended.
42149
42432
  * @example {
42150
42433
  * "video_url": "https://storage.googleapis.com/falserverless/web-examples/wan/t2v.mp4",
42151
- * "start_frame_num": 0,
42152
42434
  * "reverse_video": false,
42435
+ * "start_frame_num": 0,
42153
42436
  * "limit_num_frames": false,
42154
42437
  * "resample_fps": false,
42155
42438
  * "strength": 1,
@@ -62127,9 +62410,9 @@ export interface FooocusUpscaleOrVaryInput {
62127
62410
  * Styles
62128
62411
  * @description The style to use.
62129
62412
  * @default [
62130
- * "Fooocus Sharp",
62413
+ * "Fooocus Enhance",
62131
62414
  * "Fooocus V2",
62132
- * "Fooocus Enhance"
62415
+ * "Fooocus Sharp"
62133
62416
  * ]
62134
62417
  */
62135
62418
  styles?: (
@@ -62636,9 +62919,9 @@ export interface FooocusInpaintInput {
62636
62919
  * Styles
62637
62920
  * @description The style to use.
62638
62921
  * @default [
62639
- * "Fooocus Sharp",
62922
+ * "Fooocus Enhance",
62640
62923
  * "Fooocus V2",
62641
- * "Fooocus Enhance"
62924
+ * "Fooocus Sharp"
62642
62925
  * ]
62643
62926
  */
62644
62927
  styles?: (
@@ -63087,9 +63370,9 @@ export interface FooocusImagePromptInput {
63087
63370
  * Styles
63088
63371
  * @description The style to use.
63089
63372
  * @default [
63090
- * "Fooocus Sharp",
63373
+ * "Fooocus Enhance",
63091
63374
  * "Fooocus V2",
63092
- * "Fooocus Enhance"
63375
+ * "Fooocus Sharp"
63093
63376
  * ]
63094
63377
  */
63095
63378
  styles?: (
@@ -63545,9 +63828,9 @@ export interface FooocusInput {
63545
63828
  * Styles
63546
63829
  * @description The style to use.
63547
63830
  * @default [
63548
- * "Fooocus Sharp",
63831
+ * "Fooocus Enhance",
63549
63832
  * "Fooocus V2",
63550
- * "Fooocus Enhance"
63833
+ * "Fooocus Sharp"
63551
63834
  * ]
63552
63835
  */
63553
63836
  styles?: (
@@ -67667,7 +67950,125 @@ export interface Flux2LoraEditInput {
67667
67950
  /**
67668
67951
  * Prompt
67669
67952
  * @description The prompt to generate an image from.
67670
- * @example Make this donut realistic
67953
+ * @example Make this donut realistic
67954
+ */
67955
+ prompt: string;
67956
+ /**
67957
+ * Seed
67958
+ * @description The seed to use for the generation. If not provided, a random seed will be used.
67959
+ */
67960
+ seed?: number;
67961
+ /**
67962
+ * Sync Mode
67963
+ * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
67964
+ * @default false
67965
+ */
67966
+ sync_mode?: boolean;
67967
+ }
67968
+
67969
+ export interface Flux2LoraEditOutput {
67970
+ /**
67971
+ * Has Nsfw Concepts
67972
+ * @description Whether the generated images contain NSFW concepts.
67973
+ */
67974
+ has_nsfw_concepts: boolean[];
67975
+ /**
67976
+ * Images
67977
+ * @description The edited images
67978
+ * @example [
67979
+ * {
67980
+ * "url": "https://storage.googleapis.com/falserverless/example_outputs/flux2_dev_lora_edit_output.png"
67981
+ * }
67982
+ * ]
67983
+ */
67984
+ images: Components.ImageFile[];
67985
+ /**
67986
+ * Prompt
67987
+ * @description The prompt used for generating the image.
67988
+ */
67989
+ prompt: string;
67990
+ /**
67991
+ * Seed
67992
+ * @description Seed of the generated Image. It will be the same value of the one passed in the
67993
+ * input or the randomly generated that was used in case none was passed.
67994
+ */
67995
+ seed: number;
67996
+ timings: {
67997
+ [key: string]: number;
67998
+ };
67999
+ }
68000
+
68001
+ export interface Flux2LoraInput {
68002
+ /**
68003
+ * Acceleration
68004
+ * @description The acceleration level to use for the image generation.
68005
+ * @default regular
68006
+ * @example regular
68007
+ * @enum {string}
68008
+ */
68009
+ acceleration?: 'none' | 'regular' | 'high';
68010
+ /**
68011
+ * Enable Prompt Expansion
68012
+ * @description If set to true, the prompt will be expanded for better results.
68013
+ * @default false
68014
+ */
68015
+ enable_prompt_expansion?: boolean;
68016
+ /**
68017
+ * Enable Safety Checker
68018
+ * @description If set to true, the safety checker will be enabled.
68019
+ * @default true
68020
+ */
68021
+ enable_safety_checker?: boolean;
68022
+ /**
68023
+ * Guidance Scale
68024
+ * @description Guidance Scale is a measure of how close you want the model to stick to your prompt when looking for a related image to show you.
68025
+ * @default 2.5
68026
+ */
68027
+ guidance_scale?: number;
68028
+ /**
68029
+ * Image Size
68030
+ * @description The size of the image to generate. The width and height must be between 512 and 2048 pixels.
68031
+ * @default landscape_4_3
68032
+ */
68033
+ image_size?:
68034
+ | Components.ImageSize
68035
+ | (
68036
+ | 'square_hd'
68037
+ | 'square'
68038
+ | 'portrait_4_3'
68039
+ | 'portrait_16_9'
68040
+ | 'landscape_4_3'
68041
+ | 'landscape_16_9'
68042
+ );
68043
+ /**
68044
+ * Loras
68045
+ * @description List of LoRA weights to apply (maximum 3). Each LoRA can be a URL, HuggingFace repo ID, or local path.
68046
+ * @default []
68047
+ */
68048
+ loras?: Components.LoRAInput[];
68049
+ /**
68050
+ * Number of Images
68051
+ * @description The number of images to generate.
68052
+ * @default 1
68053
+ */
68054
+ num_images?: number;
68055
+ /**
68056
+ * Number of Inference Steps
68057
+ * @description The number of inference steps to perform.
68058
+ * @default 28
68059
+ */
68060
+ num_inference_steps?: number;
68061
+ /**
68062
+ * Output Format
68063
+ * @description The format of the generated image.
68064
+ * @default png
68065
+ * @enum {string}
68066
+ */
68067
+ output_format?: 'jpeg' | 'png' | 'webp';
68068
+ /**
68069
+ * Prompt
68070
+ * @description The prompt to generate an image from.
68071
+ * @example Close shot a pianist plays in a luxurious room with tall windows overlooking a rainy metropolis. Shot with a 50mm lens at a side profile angle, soft tungsten light highlighting hands moving over keys. Capture detailed reflections in polished black piano surfaces, raindrops sliding down glass, and atmospheric warm/cool lighting contrast.
67671
68072
  */
67672
68073
  prompt: string;
67673
68074
  /**
@@ -67683,7 +68084,7 @@ export interface Flux2LoraEditInput {
67683
68084
  sync_mode?: boolean;
67684
68085
  }
67685
68086
 
67686
- export interface Flux2LoraEditOutput {
68087
+ export interface Flux2LoraOutput {
67687
68088
  /**
67688
68089
  * Has Nsfw Concepts
67689
68090
  * @description Whether the generated images contain NSFW concepts.
@@ -67691,10 +68092,10 @@ export interface Flux2LoraEditOutput {
67691
68092
  has_nsfw_concepts: boolean[];
67692
68093
  /**
67693
68094
  * Images
67694
- * @description The edited images
68095
+ * @description The generated images
67695
68096
  * @example [
67696
68097
  * {
67697
- * "url": "https://storage.googleapis.com/falserverless/example_outputs/flux2_dev_lora_edit_output.png"
68098
+ * "url": "https://storage.googleapis.com/falserverless/example_outputs/flux2_dev_lora_t2i_output.png"
67698
68099
  * }
67699
68100
  * ]
67700
68101
  */
@@ -67715,122 +68116,68 @@ export interface Flux2LoraEditOutput {
67715
68116
  };
67716
68117
  }
67717
68118
 
67718
- export interface Flux2LoraInput {
68119
+ export interface Flux2KleinRealtimeInput {
67719
68120
  /**
67720
- * Acceleration
67721
- * @description The acceleration level to use for the image generation.
67722
- * @default regular
67723
- * @example regular
67724
- * @enum {string}
67725
- */
67726
- acceleration?: 'none' | 'regular' | 'high';
67727
- /**
67728
- * Enable Prompt Expansion
67729
- * @description If set to true, the prompt will be expanded for better results.
68121
+ * Enable Interpolation
68122
+ * @description Enable RIFE frame interpolation between consecutive frames (doubles output frames).
67730
68123
  * @default false
67731
68124
  */
67732
- enable_prompt_expansion?: boolean;
67733
- /**
67734
- * Enable Safety Checker
67735
- * @description If set to true, the safety checker will be enabled.
67736
- * @default true
67737
- */
67738
- enable_safety_checker?: boolean;
67739
- /**
67740
- * Guidance Scale
67741
- * @description Guidance Scale is a measure of how close you want the model to stick to your prompt when looking for a related image to show you.
67742
- * @default 2.5
67743
- */
67744
- guidance_scale?: number;
68125
+ enable_interpolation?: boolean;
67745
68126
  /**
67746
68127
  * Image Size
67747
- * @description The size of the image to generate. The width and height must be between 512 and 2048 pixels.
67748
- * @default landscape_4_3
67749
- */
67750
- image_size?:
67751
- | Components.ImageSize
67752
- | (
67753
- | 'square_hd'
67754
- | 'square'
67755
- | 'portrait_4_3'
67756
- | 'portrait_16_9'
67757
- | 'landscape_4_3'
67758
- | 'landscape_16_9'
67759
- );
67760
- /**
67761
- * Loras
67762
- * @description List of LoRA weights to apply (maximum 3). Each LoRA can be a URL, HuggingFace repo ID, or local path.
67763
- * @default []
68128
+ * @description The size of the generated image. square=768x768, square_hd=1024x1024.
68129
+ * @default square
68130
+ * @enum {string}
67764
68131
  */
67765
- loras?: Components.LoRAInput[];
68132
+ image_size?: 'square' | 'square_hd';
67766
68133
  /**
67767
- * Number of Images
67768
- * @description The number of images to generate.
67769
- * @default 1
68134
+ * Image URL
68135
+ * @description Base64-encoded image data URI for editing. CDN URLs are not supported for realtime. For optimal performance, use 704x704 JPEG images with 50% quality. Other sizes will be resized automatically.
67770
68136
  */
67771
- num_images?: number;
68137
+ image_url: string;
67772
68138
  /**
67773
- * Number of Inference Steps
67774
- * @description The number of inference steps to perform.
67775
- * @default 28
68139
+ * Num Inference Steps
68140
+ * @default 3
67776
68141
  */
67777
68142
  num_inference_steps?: number;
67778
68143
  /**
67779
- * Output Format
67780
- * @description The format of the generated image.
67781
- * @default png
67782
- * @enum {string}
68144
+ * Output Feedback Strength
68145
+ * @description Output feedback loop. 1.0 = pure noise (no feedback), 0.9 = 90% noise + 10% previous output latent.
68146
+ * @default 1
67783
68147
  */
67784
- output_format?: 'jpeg' | 'png' | 'webp';
68148
+ output_feedback_strength?: number;
67785
68149
  /**
67786
68150
  * Prompt
67787
- * @description The prompt to generate an image from.
67788
- * @example Close shot a pianist plays in a luxurious room with tall windows overlooking a rainy metropolis. Shot with a 50mm lens at a side profile angle, soft tungsten light highlighting hands moving over keys. Capture detailed reflections in polished black piano surfaces, raindrops sliding down glass, and atmospheric warm/cool lighting contrast.
68151
+ * @description The prompt to guide image editing.
68152
+ * @default Turn this into "Living oil painting, melting gold and sapphire"
68153
+ * @example Turn this into a watercolor painting
67789
68154
  */
67790
- prompt: string;
68155
+ prompt?: string;
67791
68156
  /**
67792
- * Seed
67793
- * @description The seed to use for the generation. If not provided, a random seed will be used.
68157
+ * Schedule Mu
68158
+ * @description Schedule mu for time shift. 2.3=default, lower=more even denoising, 0.3=nearly linear.
68159
+ * @default 2.3
67794
68160
  */
67795
- seed?: number;
68161
+ schedule_mu?: number;
67796
68162
  /**
67797
- * Sync Mode
67798
- * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
67799
- * @default false
68163
+ * Seed
68164
+ * @description Random seed for reproducibility.
68165
+ * @default 35
67800
68166
  */
67801
- sync_mode?: boolean;
68167
+ seed?: number;
67802
68168
  }
67803
68169
 
67804
- export interface Flux2LoraOutput {
67805
- /**
67806
- * Has Nsfw Concepts
67807
- * @description Whether the generated images contain NSFW concepts.
67808
- */
67809
- has_nsfw_concepts: boolean[];
68170
+ export interface Flux2KleinRealtimeOutput {
67810
68171
  /**
67811
68172
  * Images
67812
- * @description The generated images
67813
- * @example [
67814
- * {
67815
- * "url": "https://storage.googleapis.com/falserverless/example_outputs/flux2_dev_lora_t2i_output.png"
67816
- * }
67817
- * ]
67818
- */
67819
- images: Components.ImageFile[];
67820
- /**
67821
- * Prompt
67822
- * @description The prompt used for generating the image.
68173
+ * @description Generated images as raw bytes. When interpolation is enabled, returns [interpolated_frame, current_frame] in chronological order. Otherwise returns [current_frame].
67823
68174
  */
67824
- prompt: string;
68175
+ images: Components.RawImage[];
67825
68176
  /**
67826
68177
  * Seed
67827
- * @description Seed of the generated Image. It will be the same value of the one passed in the
67828
- * input or the randomly generated that was used in case none was passed.
68178
+ * @description Seed used for generation.
67829
68179
  */
67830
68180
  seed: number;
67831
- timings: {
67832
- [key: string]: number;
67833
- };
67834
68181
  }
67835
68182
 
67836
68183
  export interface Flux2Klein9bLoraInput {
@@ -72700,6 +73047,7 @@ export interface FastSdxlInpaintingInput {
72700
73047
  * Seed
72701
73048
  * @description The same seed and the same prompt given to the same version of Stable Diffusion
72702
73049
  * will output the same image every time.
73050
+ * @default null
72703
73051
  */
72704
73052
  seed?: number;
72705
73053
  /**
@@ -72840,6 +73188,7 @@ export interface FastSdxlImageToImageInput {
72840
73188
  * Seed
72841
73189
  * @description The same seed and the same prompt given to the same version of Stable Diffusion
72842
73190
  * will output the same image every time.
73191
+ * @default null
72843
73192
  */
72844
73193
  seed?: number;
72845
73194
  /**
@@ -73287,6 +73636,7 @@ export interface FastSdxlInput {
73287
73636
  * Seed
73288
73637
  * @description The same seed and the same prompt given to the same version of Stable Diffusion
73289
73638
  * will output the same image every time.
73639
+ * @default null
73290
73640
  */
73291
73641
  seed?: number;
73292
73642
  /**
@@ -73604,6 +73954,7 @@ export interface FastLightningSdxlInput {
73604
73954
  * Seed
73605
73955
  * @description The same seed and the same prompt given to the same version of Stable Diffusion
73606
73956
  * will output the same image every time.
73957
+ * @default null
73607
73958
  */
73608
73959
  seed?: number;
73609
73960
  /**
@@ -75489,220 +75840,220 @@ export interface ElevenlabsSpeechToTextScribeV2Output {
75489
75840
  * Words
75490
75841
  * @description Word-level transcription details
75491
75842
  * @example {
75492
- * "text": "Hey,",
75843
+ * "end": 0.539,
75493
75844
  * "start": 0.079,
75494
75845
  * "type": "word",
75495
- * "end": 0.539,
75846
+ * "text": "Hey,",
75496
75847
  * "speaker_id": "speaker_0"
75497
75848
  * }
75498
75849
  * @example {
75499
- * "text": " ",
75850
+ * "end": 0.599,
75500
75851
  * "start": 0.539,
75501
75852
  * "type": "spacing",
75502
- * "end": 0.599,
75853
+ * "text": " ",
75503
75854
  * "speaker_id": "speaker_0"
75504
75855
  * }
75505
75856
  * @example {
75506
- * "text": "this",
75857
+ * "end": 0.679,
75507
75858
  * "start": 0.599,
75508
75859
  * "type": "word",
75509
- * "end": 0.679,
75860
+ * "text": "this",
75510
75861
  * "speaker_id": "speaker_0"
75511
75862
  * }
75512
75863
  * @example {
75513
- * "text": " ",
75864
+ * "end": 0.739,
75514
75865
  * "start": 0.679,
75515
75866
  * "type": "spacing",
75516
- * "end": 0.739,
75867
+ * "text": " ",
75517
75868
  * "speaker_id": "speaker_0"
75518
75869
  * }
75519
75870
  * @example {
75520
- * "text": "is",
75871
+ * "end": 0.799,
75521
75872
  * "start": 0.739,
75522
75873
  * "type": "word",
75523
- * "end": 0.799,
75874
+ * "text": "is",
75524
75875
  * "speaker_id": "speaker_0"
75525
75876
  * }
75526
75877
  * @example {
75527
- * "text": " ",
75878
+ * "end": 0.939,
75528
75879
  * "start": 0.799,
75529
75880
  * "type": "spacing",
75530
- * "end": 0.939,
75881
+ * "text": " ",
75531
75882
  * "speaker_id": "speaker_0"
75532
75883
  * }
75533
75884
  * @example {
75534
- * "text": "a",
75885
+ * "end": 0.939,
75535
75886
  * "start": 0.939,
75536
75887
  * "type": "word",
75537
- * "end": 0.939,
75888
+ * "text": "a",
75538
75889
  * "speaker_id": "speaker_0"
75539
75890
  * }
75540
75891
  * @example {
75541
- * "text": " ",
75892
+ * "end": 0.959,
75542
75893
  * "start": 0.939,
75543
75894
  * "type": "spacing",
75544
- * "end": 0.959,
75895
+ * "text": " ",
75545
75896
  * "speaker_id": "speaker_0"
75546
75897
  * }
75547
75898
  * @example {
75548
- * "text": "test",
75899
+ * "end": 1.179,
75549
75900
  * "start": 0.959,
75550
75901
  * "type": "word",
75551
- * "end": 1.179,
75902
+ * "text": "test",
75552
75903
  * "speaker_id": "speaker_0"
75553
75904
  * }
75554
75905
  * @example {
75555
- * "text": " ",
75906
+ * "end": 1.219,
75556
75907
  * "start": 1.179,
75557
75908
  * "type": "spacing",
75558
- * "end": 1.219,
75909
+ * "text": " ",
75559
75910
  * "speaker_id": "speaker_0"
75560
75911
  * }
75561
75912
  * @example {
75562
- * "text": "recording",
75913
+ * "end": 1.719,
75563
75914
  * "start": 1.22,
75564
75915
  * "type": "word",
75565
- * "end": 1.719,
75916
+ * "text": "recording",
75566
75917
  * "speaker_id": "speaker_0"
75567
75918
  * }
75568
75919
  * @example {
75569
- * "text": " ",
75920
+ * "end": 1.719,
75570
75921
  * "start": 1.719,
75571
75922
  * "type": "spacing",
75572
- * "end": 1.719,
75923
+ * "text": " ",
75573
75924
  * "speaker_id": "speaker_0"
75574
75925
  * }
75575
75926
  * @example {
75576
- * "text": "for",
75927
+ * "end": 1.86,
75577
75928
  * "start": 1.719,
75578
75929
  * "type": "word",
75579
- * "end": 1.86,
75930
+ * "text": "for",
75580
75931
  * "speaker_id": "speaker_0"
75581
75932
  * }
75582
75933
  * @example {
75583
- * "text": " ",
75934
+ * "end": 1.879,
75584
75935
  * "start": 1.86,
75585
75936
  * "type": "spacing",
75586
- * "end": 1.879,
75937
+ * "text": " ",
75587
75938
  * "speaker_id": "speaker_0"
75588
75939
  * }
75589
75940
  * @example {
75590
- * "text": "Scribe",
75941
+ * "end": 2.24,
75591
75942
  * "start": 1.879,
75592
75943
  * "type": "word",
75593
- * "end": 2.24,
75944
+ * "text": "Scribe",
75594
75945
  * "speaker_id": "speaker_0"
75595
75946
  * }
75596
75947
  * @example {
75597
- * "text": " ",
75948
+ * "end": 2.319,
75598
75949
  * "start": 2.24,
75599
75950
  * "type": "spacing",
75600
- * "end": 2.319,
75951
+ * "text": " ",
75601
75952
  * "speaker_id": "speaker_0"
75602
75953
  * }
75603
75954
  * @example {
75604
- * "text": "version",
75955
+ * "end": 2.759,
75605
75956
  * "start": 2.319,
75606
75957
  * "type": "word",
75607
- * "end": 2.759,
75958
+ * "text": "version",
75608
75959
  * "speaker_id": "speaker_0"
75609
75960
  * }
75610
75961
  * @example {
75611
- * "text": " ",
75962
+ * "end": 2.779,
75612
75963
  * "start": 2.759,
75613
75964
  * "type": "spacing",
75614
- * "end": 2.779,
75965
+ * "text": " ",
75615
75966
  * "speaker_id": "speaker_0"
75616
75967
  * }
75617
75968
  * @example {
75618
- * "text": "two,",
75969
+ * "end": 3.379,
75619
75970
  * "start": 2.779,
75620
75971
  * "type": "word",
75621
- * "end": 3.379,
75972
+ * "text": "two,",
75622
75973
  * "speaker_id": "speaker_0"
75623
75974
  * }
75624
75975
  * @example {
75625
- * "text": " ",
75976
+ * "end": 3.399,
75626
75977
  * "start": 3.379,
75627
75978
  * "type": "spacing",
75628
- * "end": 3.399,
75979
+ * "text": " ",
75629
75980
  * "speaker_id": "speaker_0"
75630
75981
  * }
75631
75982
  * @example {
75632
- * "text": "which",
75983
+ * "end": 3.519,
75633
75984
  * "start": 3.399,
75634
75985
  * "type": "word",
75635
- * "end": 3.519,
75986
+ * "text": "which",
75636
75987
  * "speaker_id": "speaker_0"
75637
75988
  * }
75638
75989
  * @example {
75639
- * "text": " ",
75990
+ * "end": 3.539,
75640
75991
  * "start": 3.519,
75641
75992
  * "type": "spacing",
75642
- * "end": 3.539,
75993
+ * "text": " ",
75643
75994
  * "speaker_id": "speaker_0"
75644
75995
  * }
75645
75996
  * @example {
75646
- * "text": "is",
75997
+ * "end": 3.659,
75647
75998
  * "start": 3.539,
75648
75999
  * "type": "word",
75649
- * "end": 3.659,
76000
+ * "text": "is",
75650
76001
  * "speaker_id": "speaker_0"
75651
76002
  * }
75652
76003
  * @example {
75653
- * "text": " ",
76004
+ * "end": 3.699,
75654
76005
  * "start": 3.659,
75655
76006
  * "type": "spacing",
75656
- * "end": 3.699,
76007
+ * "text": " ",
75657
76008
  * "speaker_id": "speaker_0"
75658
76009
  * }
75659
76010
  * @example {
75660
- * "text": "now",
76011
+ * "end": 3.839,
75661
76012
  * "start": 3.699,
75662
76013
  * "type": "word",
75663
- * "end": 3.839,
76014
+ * "text": "now",
75664
76015
  * "speaker_id": "speaker_0"
75665
76016
  * }
75666
76017
  * @example {
75667
- * "text": " ",
76018
+ * "end": 3.839,
75668
76019
  * "start": 3.839,
75669
76020
  * "type": "spacing",
75670
- * "end": 3.839,
76021
+ * "text": " ",
75671
76022
  * "speaker_id": "speaker_0"
75672
76023
  * }
75673
76024
  * @example {
75674
- * "text": "available",
76025
+ * "end": 4.319,
75675
76026
  * "start": 3.839,
75676
76027
  * "type": "word",
75677
- * "end": 4.319,
76028
+ * "text": "available",
75678
76029
  * "speaker_id": "speaker_0"
75679
76030
  * }
75680
76031
  * @example {
75681
- * "text": " ",
76032
+ * "end": 4.339,
75682
76033
  * "start": 4.319,
75683
76034
  * "type": "spacing",
75684
- * "end": 4.339,
76035
+ * "text": " ",
75685
76036
  * "speaker_id": "speaker_0"
75686
76037
  * }
75687
76038
  * @example {
75688
- * "text": "on",
76039
+ * "end": 4.579,
75689
76040
  * "start": 4.339,
75690
76041
  * "type": "word",
75691
- * "end": 4.579,
76042
+ * "text": "on",
75692
76043
  * "speaker_id": "speaker_0"
75693
76044
  * }
75694
76045
  * @example {
75695
- * "text": " ",
76046
+ * "end": 4.599,
75696
76047
  * "start": 4.579,
75697
76048
  * "type": "spacing",
75698
- * "end": 4.599,
76049
+ * "text": " ",
75699
76050
  * "speaker_id": "speaker_0"
75700
76051
  * }
75701
76052
  * @example {
75702
- * "text": "fal.ai.",
76053
+ * "end": 5.699,
75703
76054
  * "start": 4.599,
75704
76055
  * "type": "word",
75705
- * "end": 5.699,
76056
+ * "text": "fal.ai.",
75706
76057
  * "speaker_id": "speaker_0"
75707
76058
  * }
75708
76059
  */
@@ -80223,6 +80574,84 @@ export interface BytedanceSeedanceV15ProImageToVideoOutput {
80223
80574
  video: Components.File;
80224
80575
  }
80225
80576
 
80577
+ export interface BytedanceSeedV2MiniInput {
80578
+ /**
80579
+ * Image Urls
80580
+ * @description URLs of images for visual understanding. Supported formats: JPEG, PNG, WebP. A maximum of 6 images is supported. Any additional images will be ignored.
80581
+ */
80582
+ image_urls?: string[];
80583
+ /**
80584
+ * Max Completion Tokens
80585
+ * @description Controls the maximum length of the model's output, including both the model's response and its chain-of-thought content, measured in tokens.
80586
+ * @default 4096
80587
+ */
80588
+ max_completion_tokens?: number;
80589
+ /**
80590
+ * Messages
80591
+ * @description Optional prior conversation history for multi-turn conversations. Pass back the `messages` field from a previous response to provide context. The current `prompt`, `image_urls`, `video_urls`, and `system_prompt` are always appended as the latest user turn.
80592
+ * @example []
80593
+ */
80594
+ messages?: Components.Seed2MiniMessage[];
80595
+ /**
80596
+ * Prompt
80597
+ * @description The text prompt or question for the model.
80598
+ * @example What can you do?
80599
+ */
80600
+ prompt: string;
80601
+ /**
80602
+ * Reasoning Effort
80603
+ * @description Controls the depth of reasoning before the model responds. Only applicable when `thinking` is `enabled` or `auto`. `minimal` for immediate response, `low` for faster response with light reasoning, `medium` for balanced speed and depth, `high` for deep analysis of complex issues.
80604
+ */
80605
+ reasoning_effort?: 'minimal' | 'low' | 'medium' | 'high';
80606
+ /**
80607
+ * System Prompt
80608
+ * @description Optional system prompt to guide the model's behavior.
80609
+ */
80610
+ system_prompt?: string;
80611
+ /**
80612
+ * Temperature
80613
+ * @description Controls randomness in the response. Lower values make output more focused and deterministic, higher values make it more creative.
80614
+ * @default 1
80615
+ */
80616
+ temperature?: number;
80617
+ /**
80618
+ * Thinking
80619
+ * @description Controls the model's chain-of-thought reasoning. `enabled` always includes reasoning, `disabled` never includes reasoning, `auto` lets the model decide based on the query.
80620
+ * @default enabled
80621
+ * @enum {string}
80622
+ */
80623
+ thinking?: 'enabled' | 'disabled' | 'auto';
80624
+ /**
80625
+ * Top P
80626
+ * @description Nucleus sampling parameter. The model considers tokens with top_p cumulative probability mass. Lower values narrow the token selection.
80627
+ * @default 0.7
80628
+ */
80629
+ top_p?: number;
80630
+ /**
80631
+ * Video Urls
80632
+ * @description URLs of videos for video understanding. Supported formats: MP4, MOV. Audio comprehension is not supported. A maximum of 3 videos is supported. Any additional videos will be ignored.
80633
+ */
80634
+ video_urls?: string[];
80635
+ }
80636
+
80637
+ export interface BytedanceSeedV2MiniOutput {
80638
+ /**
80639
+ * Messages
80640
+ * @description The full conversation history including the model's response. Pass this back as the `messages` input field to continue the conversation.
80641
+ */
80642
+ messages: Components.Seed2MiniMessage[];
80643
+ /**
80644
+ * Output
80645
+ * @description The model's text response.
80646
+ */
80647
+ output: string;
80648
+ /**
80649
+ * Reasoning Content
80650
+ * @description The model's chain-of-thought reasoning content. Only present when `thinking` is `enabled` or `auto`.
80651
+ */
80652
+ reasoning_content?: string;
80653
+ }
80654
+
80226
80655
  export interface BytedanceOmnihumanV15Input {
80227
80656
  /**
80228
80657
  * Audio Url
@@ -81486,8 +81915,8 @@ export interface BagelEditOutput {
81486
81915
  * @description The edited images.
81487
81916
  * @example [
81488
81917
  * {
81489
- * "file_size": 423052,
81490
81918
  * "height": 1024,
81919
+ * "file_size": 423052,
81491
81920
  * "file_name": "hQnndOMvGSt2UsYAiV3vs.jpeg",
81492
81921
  * "content_type": "image/jpeg",
81493
81922
  * "url": "https://storage.googleapis.com/falserverless/bagel/hQnndOMvGSt2UsYAiV3vs.jpeg",
@@ -81549,8 +81978,8 @@ export interface BagelOutput {
81549
81978
  * @description The generated images.
81550
81979
  * @example [
81551
81980
  * {
81552
- * "file_size": 423052,
81553
81981
  * "height": 1024,
81982
+ * "file_size": 423052,
81554
81983
  * "file_name": "wRhCPSyiKTiLnnWvUpGIl.jpeg",
81555
81984
  * "content_type": "image/jpeg",
81556
81985
  * "url": "https://storage.googleapis.com/falserverless/bagel/wRhCPSyiKTiLnnWvUpGIl.jpeg",
@@ -83153,8 +83582,8 @@ export interface ClarityaiCrystalVideoUpscalerOutput {
83153
83582
  * "height": 2160,
83154
83583
  * "duration": 13.056527,
83155
83584
  * "url": "https://storage.googleapis.com/falserverless/example_outputs/crystal_upscaler/video_upscaling/video_out.mp4",
83156
- * "width": 4096,
83157
83585
  * "fps": 23.130193905817176,
83586
+ * "width": 4096,
83158
83587
  * "file_name": "w0VQQvPdwvV2GSCtRTMzh_hDH8SPrB.mp4",
83159
83588
  * "content_type": "video/mp4",
83160
83589
  * "num_frames": 302