fal-endpoint-types 1.3.31 → 1.3.33

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.
@@ -1831,100 +1831,6 @@ export interface SharedType_faf {
1831
1831
  sync_mode?: boolean;
1832
1832
  }
1833
1833
 
1834
- export interface SharedType_fa0 {
1835
- /**
1836
- * Control Lora Image Url
1837
- * @description The image to use for control lora. This is used to control the style of the generated image.
1838
- * @example https://v3.fal.media/files/kangaroo/Cb7BeM7G4DauK_lWjzY3N_Celeb6.jpg
1839
- */
1840
- control_lora_image_url: string;
1841
- /**
1842
- * Control Lora Strength
1843
- * @description The strength of the control lora.
1844
- * @default 1
1845
- */
1846
- control_lora_strength?: number;
1847
- /**
1848
- * Enable Safety Checker
1849
- * @description If set to true, the safety checker will be enabled.
1850
- * @default true
1851
- */
1852
- enable_safety_checker?: boolean;
1853
- /**
1854
- * Guidance scale (CFG)
1855
- * @description The CFG (Classifier Free Guidance) scale is a measure of how close you want
1856
- * the model to stick to your prompt when looking for a related image to show you.
1857
- * @default 3.5
1858
- */
1859
- guidance_scale?: number;
1860
- /**
1861
- * Image Size
1862
- * @description The size of the generated image.
1863
- * @default landscape_4_3
1864
- */
1865
- image_size?:
1866
- | Components.ImageSize
1867
- | (
1868
- | 'square_hd'
1869
- | 'square'
1870
- | 'portrait_4_3'
1871
- | 'portrait_16_9'
1872
- | 'landscape_4_3'
1873
- | 'landscape_16_9'
1874
- );
1875
- /**
1876
- * Loras
1877
- * @description The LoRAs to use for the image generation. You can use any number of LoRAs
1878
- * and they will be merged together to generate the final image.
1879
- * @default []
1880
- */
1881
- loras?: Components.LoraWeight[];
1882
- /**
1883
- * Num Images
1884
- * @description The number of images to generate.
1885
- * @default 1
1886
- */
1887
- num_images?: number;
1888
- /**
1889
- * Num Inference Steps
1890
- * @description The number of inference steps to perform.
1891
- * @default 28
1892
- */
1893
- num_inference_steps?: number;
1894
- /**
1895
- * Output Format
1896
- * @description The format of the generated image.
1897
- * @default jpeg
1898
- * @enum {string}
1899
- */
1900
- output_format?: 'jpeg' | 'png';
1901
- /**
1902
- * Preprocess Depth
1903
- * @description If set to true, the input image will be preprocessed to extract depth information.
1904
- * This is useful for generating depth maps from images.
1905
- * @default true
1906
- */
1907
- preprocess_depth?: boolean;
1908
- /**
1909
- * Prompt
1910
- * @description The prompt to generate an image from.
1911
- * @example Extreme close-up of a single tiger eye, direct frontal view. Detailed iris and pupil. Sharp focus on eye texture and color. Natural lighting to capture authentic eye shine and depth. The word "FLUX" is painted over it in big, white brush strokes with visible texture.
1912
- */
1913
- prompt: string;
1914
- /**
1915
- * Seed
1916
- * @description The same seed and the same prompt given to the same version of the model
1917
- * will output the same image every time.
1918
- */
1919
- seed?: number;
1920
- /**
1921
- * Sync Mode
1922
- * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
1923
- * @default false
1924
- */
1925
- sync_mode?: boolean;
1926
- }
1927
-
1928
1834
  export interface SharedType_f7a {
1929
1835
  config_file: Components.File;
1930
1836
  debug_dataset?: Components.File;
@@ -3423,6 +3329,90 @@ export interface SharedType_d66 {
3423
3329
  voice_setting?: Components.VoiceSetting;
3424
3330
  }
3425
3331
 
3332
+ export interface SharedType_d53 {
3333
+ animation_fbx?: Components.File;
3334
+ animation_glb?: Components.File;
3335
+ basic_animations?: Components.BasicAnimations;
3336
+ /**
3337
+ * @description Generated 3D object in GLB format.
3338
+ * @example {
3339
+ * "file_size": 9242744,
3340
+ * "file_name": "model.glb",
3341
+ * "content_type": "model/gltf-binary",
3342
+ * "url": "https://v3b.fal.media/files/b/zebra/OXF1e1bO3JddPTaugv0eL_model.glb"
3343
+ * }
3344
+ */
3345
+ model_glb: Components.File;
3346
+ /**
3347
+ * @description URLs for different 3D model formats
3348
+ * @example {
3349
+ * "fbx": {
3350
+ * "file_size": 5427052,
3351
+ * "file_name": "model.fbx",
3352
+ * "content_type": "application/octet-stream",
3353
+ * "url": "https://v3b.fal.media/files/b/kangaroo/4Q2qdpTvfLVdzAKH1-72v_model.fbx"
3354
+ * },
3355
+ * "usdz": {
3356
+ * "file_size": 9991969,
3357
+ * "file_name": "model.usdz",
3358
+ * "content_type": "model/vnd.usdz+zip",
3359
+ * "url": "https://v3b.fal.media/files/b/lion/RgJG9EBQ_GAHMVWV3wCis_model.usdz"
3360
+ * },
3361
+ * "obj": {
3362
+ * "file_size": 2744413,
3363
+ * "file_name": "model.obj",
3364
+ * "content_type": "text/plain",
3365
+ * "url": "https://v3b.fal.media/files/b/koala/_Vg0d084-hd3EdpIJDf7U_model.obj"
3366
+ * },
3367
+ * "glb": {
3368
+ * "file_size": 9242744,
3369
+ * "file_name": "model.glb",
3370
+ * "content_type": "model/gltf-binary",
3371
+ * "url": "https://v3b.fal.media/files/b/zebra/OXF1e1bO3JddPTaugv0eL_model.glb"
3372
+ * }
3373
+ * }
3374
+ */
3375
+ model_urls: Components.ModelUrls;
3376
+ /**
3377
+ * Rig Task Id
3378
+ * @description Rigging task ID. Only present when enable_rigging is true.
3379
+ */
3380
+ rig_task_id?: string;
3381
+ rigged_character_fbx?: Components.File;
3382
+ rigged_character_glb?: Components.File;
3383
+ /**
3384
+ * Seed
3385
+ * @description The seed used for generation (if available)
3386
+ * @example 2009275957
3387
+ */
3388
+ seed?: number;
3389
+ /**
3390
+ * Texture Urls
3391
+ * @description Array of texture file objects, matching Meshy API structure
3392
+ * @example [
3393
+ * {
3394
+ * "base_color": {
3395
+ * "file_size": 4328755,
3396
+ * "file_name": "texture_0.png",
3397
+ * "content_type": "image/png",
3398
+ * "url": "https://v3b.fal.media/files/b/tiger/NkgxcEom_42V4_8UUXiRR_texture_0.png"
3399
+ * }
3400
+ * }
3401
+ * ]
3402
+ */
3403
+ texture_urls?: Components.TextureFiles[];
3404
+ /**
3405
+ * @description Preview thumbnail of the generated model
3406
+ * @example {
3407
+ * "file_size": 54279,
3408
+ * "file_name": "preview.png",
3409
+ * "content_type": "image/png",
3410
+ * "url": "https://v3b.fal.media/files/b/penguin/rfnS6ClmeEWgDXp_oD5tN_preview.png"
3411
+ * }
3412
+ */
3413
+ thumbnail?: Components.File;
3414
+ }
3415
+
3426
3416
  export interface SharedType_d35 {
3427
3417
  /**
3428
3418
  * Audio Url
@@ -4613,101 +4603,6 @@ export interface SharedType_ba2 {
4613
4603
  sync_mode?: boolean;
4614
4604
  }
4615
4605
 
4616
- export interface SharedType_ba1 {
4617
- /**
4618
- * Actual Prompt
4619
- * @description The actual prompt used if prompt expansion was enabled
4620
- */
4621
- actual_prompt?: string;
4622
- animation_fbx?: Components.File;
4623
- animation_glb?: Components.File;
4624
- basic_animations?: Components.BasicAnimations;
4625
- /**
4626
- * @description Generated 3D object in GLB format.
4627
- * @example {
4628
- * "file_size": 9314028,
4629
- * "file_name": "model.glb",
4630
- * "content_type": "model/gltf-binary",
4631
- * "url": "https://v3b.fal.media/files/b/penguin/DId89qXLu6BXu09RFAwAV_model.glb"
4632
- * }
4633
- */
4634
- model_glb: Components.File;
4635
- /**
4636
- * @description URLs for different 3D model formats
4637
- * @example {
4638
- * "fbx": {
4639
- * "file_size": 5444380,
4640
- * "file_name": "model.fbx",
4641
- * "content_type": "application/octet-stream",
4642
- * "url": "https://v3b.fal.media/files/b/kangaroo/7nUUw5dHN9a0DKlOpAKbP_model.fbx"
4643
- * },
4644
- * "usdz": {
4645
- * "file_size": 9834246,
4646
- * "file_name": "model.usdz",
4647
- * "content_type": "model/vnd.usdz+zip",
4648
- * "url": "https://v3b.fal.media/files/b/panda/XcC-mIJywUvH7coyrzENU_model.usdz"
4649
- * },
4650
- * "glb": {
4651
- * "file_size": 9314028,
4652
- * "file_name": "model.glb",
4653
- * "content_type": "model/gltf-binary",
4654
- * "url": "https://v3b.fal.media/files/b/penguin/DId89qXLu6BXu09RFAwAV_model.glb"
4655
- * },
4656
- * "obj": {
4657
- * "file_size": 2755145,
4658
- * "file_name": "model.obj",
4659
- * "content_type": "text/plain",
4660
- * "url": "https://v3b.fal.media/files/b/monkey/cCNMHqUbKSNtDN1iGmiYm_model.obj"
4661
- * }
4662
- * }
4663
- */
4664
- model_urls: Components.ModelUrls;
4665
- /**
4666
- * Prompt
4667
- * @description The text prompt used for generation
4668
- * @example A rustic, antique wooden treasure chest with a curved, domed lid, constructed from weathered, dark brown planks exhibiting prominent wood grain and subtle distress. It's heavily reinforced with broad, dark grey, oxidized metal bands secured by numerous circular rivets. Ornate, dark iron decorative elements featuring swirling foliate patterns and dragon motifs adorn the corners and lid. A prominent, circular, intricately carved metal lock plate with a central keyhole dominates the front, flanked by two large, dark metallic pull rings.
4669
- */
4670
- prompt: string;
4671
- /**
4672
- * Rig Task Id
4673
- * @description Rigging task ID. Only present when enable_rigging is true.
4674
- */
4675
- rig_task_id?: string;
4676
- rigged_character_fbx?: Components.File;
4677
- rigged_character_glb?: Components.File;
4678
- /**
4679
- * Seed
4680
- * @description The seed used for generation
4681
- * @example 4002110719
4682
- */
4683
- seed?: number;
4684
- /**
4685
- * Texture Urls
4686
- * @description Array of texture file objects
4687
- * @example [
4688
- * {
4689
- * "base_color": {
4690
- * "file_size": 4254502,
4691
- * "file_name": "texture_0.png",
4692
- * "content_type": "image/png",
4693
- * "url": "https://v3b.fal.media/files/b/panda/DoPKAuZY0tTjnr6C9ee-Q_texture_0.png"
4694
- * }
4695
- * }
4696
- * ]
4697
- */
4698
- texture_urls?: Components.TextureFiles[];
4699
- /**
4700
- * @description Preview thumbnail of the generated model
4701
- * @example {
4702
- * "file_size": 173792,
4703
- * "file_name": "preview.png",
4704
- * "content_type": "image/png",
4705
- * "url": "https://v3b.fal.media/files/b/koala/6LJISu4ilkZXcdOETwl_d_preview.png"
4706
- * }
4707
- */
4708
- thumbnail?: Components.File;
4709
- }
4710
-
4711
4606
  export interface SharedType_b89 {
4712
4607
  /**
4713
4608
  * Description
@@ -6379,6 +6274,105 @@ export interface SharedType_913 {
6379
6274
  video_write_mode?: 'fast' | 'balanced' | 'small';
6380
6275
  }
6381
6276
 
6277
+ export interface SharedType_90c {
6278
+ /**
6279
+ * Acceleration
6280
+ * @description Acceleration level for image generation. 'regular' balances speed and quality.
6281
+ * @default none
6282
+ * @enum {string}
6283
+ */
6284
+ acceleration?: 'none' | 'regular';
6285
+ /**
6286
+ * Enable Safety Checker
6287
+ * @description If set to true, the safety checker will be enabled.
6288
+ * @default true
6289
+ */
6290
+ enable_safety_checker?: boolean;
6291
+ /**
6292
+ * Guidance scale (CFG)
6293
+ * @description The CFG (Classifier Free Guidance) scale is a measure of how close you want
6294
+ * the model to stick to your prompt when looking for a related image to show you.
6295
+ * @default 3.5
6296
+ */
6297
+ guidance_scale?: number;
6298
+ /**
6299
+ * Image Size
6300
+ * @description The size of the generated image.
6301
+ */
6302
+ image_size?:
6303
+ | Components.ImageSize
6304
+ | (
6305
+ | 'square_hd'
6306
+ | 'square'
6307
+ | 'portrait_4_3'
6308
+ | 'portrait_16_9'
6309
+ | 'landscape_4_3'
6310
+ | 'landscape_16_9'
6311
+ );
6312
+ /**
6313
+ * Image Url
6314
+ * @description URL of image to use for inpainting. or img2img
6315
+ * @example https://storage.googleapis.com/falserverless/example_inputs/dog.png
6316
+ */
6317
+ image_url: string;
6318
+ /**
6319
+ * Loras
6320
+ * @description The LoRAs to use for the image generation. You can use any number of LoRAs
6321
+ * and they will be merged together to generate the final image.
6322
+ * @default []
6323
+ */
6324
+ loras?: Components.LoraWeight[];
6325
+ /**
6326
+ * Mask Url
6327
+ * @description The mask to area to Inpaint in.
6328
+ * @example https://storage.googleapis.com/falserverless/example_inputs/dog_mask.png
6329
+ */
6330
+ mask_url: string;
6331
+ /**
6332
+ * Num Images
6333
+ * @description The number of images to generate. This is always set to 1 for streaming output.
6334
+ * @default 1
6335
+ */
6336
+ num_images?: number;
6337
+ /**
6338
+ * Num Inference Steps
6339
+ * @description The number of inference steps to perform.
6340
+ * @default 28
6341
+ */
6342
+ num_inference_steps?: number;
6343
+ /**
6344
+ * Output Format
6345
+ * @description The format of the generated image.
6346
+ * @default jpeg
6347
+ * @enum {string}
6348
+ */
6349
+ output_format?: 'jpeg' | 'png';
6350
+ /**
6351
+ * Prompt
6352
+ * @description The prompt to generate an image from.
6353
+ * @example A photo of a lion sitting on a stone bench
6354
+ */
6355
+ prompt: string;
6356
+ /**
6357
+ * Seed
6358
+ * @description The same seed and the same prompt given to the same version of the model
6359
+ * will output the same image every time.
6360
+ */
6361
+ seed?: number;
6362
+ /**
6363
+ * Strength
6364
+ * @description The strength to use for inpainting/image-to-image. Only used if the image_url is provided. 1.0 is completely remakes the image while 0.0 preserves the original.
6365
+ * @default 0.85
6366
+ */
6367
+ strength?: number;
6368
+ /**
6369
+ * Sync Mode
6370
+ * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
6371
+ * @default false
6372
+ */
6373
+ sync_mode?: boolean;
6374
+ }
6375
+
6382
6376
  export interface SharedType_905 {
6383
6377
  /**
6384
6378
  * Aspect Ratio
@@ -7112,90 +7106,6 @@ export interface SharedType_82f {
7112
7106
  video_url: string;
7113
7107
  }
7114
7108
 
7115
- export interface SharedType_82d {
7116
- animation_fbx?: Components.File;
7117
- animation_glb?: Components.File;
7118
- basic_animations?: Components.BasicAnimations;
7119
- /**
7120
- * @description Generated 3D object in GLB format.
7121
- * @example {
7122
- * "file_size": 9242744,
7123
- * "file_name": "model.glb",
7124
- * "content_type": "model/gltf-binary",
7125
- * "url": "https://v3b.fal.media/files/b/zebra/OXF1e1bO3JddPTaugv0eL_model.glb"
7126
- * }
7127
- */
7128
- model_glb: Components.File;
7129
- /**
7130
- * @description URLs for different 3D model formats
7131
- * @example {
7132
- * "fbx": {
7133
- * "file_size": 5427052,
7134
- * "file_name": "model.fbx",
7135
- * "content_type": "application/octet-stream",
7136
- * "url": "https://v3b.fal.media/files/b/kangaroo/4Q2qdpTvfLVdzAKH1-72v_model.fbx"
7137
- * },
7138
- * "usdz": {
7139
- * "file_size": 9991969,
7140
- * "file_name": "model.usdz",
7141
- * "content_type": "model/vnd.usdz+zip",
7142
- * "url": "https://v3b.fal.media/files/b/lion/RgJG9EBQ_GAHMVWV3wCis_model.usdz"
7143
- * },
7144
- * "glb": {
7145
- * "file_size": 9242744,
7146
- * "file_name": "model.glb",
7147
- * "content_type": "model/gltf-binary",
7148
- * "url": "https://v3b.fal.media/files/b/zebra/OXF1e1bO3JddPTaugv0eL_model.glb"
7149
- * },
7150
- * "obj": {
7151
- * "file_size": 2744413,
7152
- * "file_name": "model.obj",
7153
- * "content_type": "text/plain",
7154
- * "url": "https://v3b.fal.media/files/b/koala/_Vg0d084-hd3EdpIJDf7U_model.obj"
7155
- * }
7156
- * }
7157
- */
7158
- model_urls: Components.ModelUrls;
7159
- /**
7160
- * Rig Task Id
7161
- * @description Rigging task ID. Only present when enable_rigging is true.
7162
- */
7163
- rig_task_id?: string;
7164
- rigged_character_fbx?: Components.File;
7165
- rigged_character_glb?: Components.File;
7166
- /**
7167
- * Seed
7168
- * @description The seed used for generation (if available)
7169
- * @example 2009275957
7170
- */
7171
- seed?: number;
7172
- /**
7173
- * Texture Urls
7174
- * @description Array of texture file objects, matching Meshy API structure
7175
- * @example [
7176
- * {
7177
- * "base_color": {
7178
- * "file_size": 4328755,
7179
- * "file_name": "texture_0.png",
7180
- * "content_type": "image/png",
7181
- * "url": "https://v3b.fal.media/files/b/tiger/NkgxcEom_42V4_8UUXiRR_texture_0.png"
7182
- * }
7183
- * }
7184
- * ]
7185
- */
7186
- texture_urls?: Components.TextureFiles[];
7187
- /**
7188
- * @description Preview thumbnail of the generated model
7189
- * @example {
7190
- * "file_size": 54279,
7191
- * "file_name": "preview.png",
7192
- * "content_type": "image/png",
7193
- * "url": "https://v3b.fal.media/files/b/penguin/rfnS6ClmeEWgDXp_oD5tN_preview.png"
7194
- * }
7195
- */
7196
- thumbnail?: Components.File;
7197
- }
7198
-
7199
7109
  export interface SharedType_82b {
7200
7110
  /**
7201
7111
  * @description The generated video
@@ -8017,6 +7927,101 @@ export interface SharedType_772 {
8017
7927
  video_url: string;
8018
7928
  }
8019
7929
 
7930
+ export interface SharedType_752 {
7931
+ /**
7932
+ * Actual Prompt
7933
+ * @description The actual prompt used if prompt expansion was enabled
7934
+ */
7935
+ actual_prompt?: string;
7936
+ animation_fbx?: Components.File;
7937
+ animation_glb?: Components.File;
7938
+ basic_animations?: Components.BasicAnimations;
7939
+ /**
7940
+ * @description Generated 3D object in GLB format.
7941
+ * @example {
7942
+ * "file_size": 9314028,
7943
+ * "file_name": "model.glb",
7944
+ * "content_type": "model/gltf-binary",
7945
+ * "url": "https://v3b.fal.media/files/b/penguin/DId89qXLu6BXu09RFAwAV_model.glb"
7946
+ * }
7947
+ */
7948
+ model_glb: Components.File;
7949
+ /**
7950
+ * @description URLs for different 3D model formats
7951
+ * @example {
7952
+ * "fbx": {
7953
+ * "file_size": 5444380,
7954
+ * "file_name": "model.fbx",
7955
+ * "content_type": "application/octet-stream",
7956
+ * "url": "https://v3b.fal.media/files/b/kangaroo/7nUUw5dHN9a0DKlOpAKbP_model.fbx"
7957
+ * },
7958
+ * "usdz": {
7959
+ * "file_size": 9834246,
7960
+ * "file_name": "model.usdz",
7961
+ * "content_type": "model/vnd.usdz+zip",
7962
+ * "url": "https://v3b.fal.media/files/b/panda/XcC-mIJywUvH7coyrzENU_model.usdz"
7963
+ * },
7964
+ * "obj": {
7965
+ * "file_size": 2755145,
7966
+ * "file_name": "model.obj",
7967
+ * "content_type": "text/plain",
7968
+ * "url": "https://v3b.fal.media/files/b/monkey/cCNMHqUbKSNtDN1iGmiYm_model.obj"
7969
+ * },
7970
+ * "glb": {
7971
+ * "file_size": 9314028,
7972
+ * "file_name": "model.glb",
7973
+ * "content_type": "model/gltf-binary",
7974
+ * "url": "https://v3b.fal.media/files/b/penguin/DId89qXLu6BXu09RFAwAV_model.glb"
7975
+ * }
7976
+ * }
7977
+ */
7978
+ model_urls: Components.ModelUrls;
7979
+ /**
7980
+ * Prompt
7981
+ * @description The text prompt used for generation
7982
+ * @example A rustic, antique wooden treasure chest with a curved, domed lid, constructed from weathered, dark brown planks exhibiting prominent wood grain and subtle distress. It's heavily reinforced with broad, dark grey, oxidized metal bands secured by numerous circular rivets. Ornate, dark iron decorative elements featuring swirling foliate patterns and dragon motifs adorn the corners and lid. A prominent, circular, intricately carved metal lock plate with a central keyhole dominates the front, flanked by two large, dark metallic pull rings.
7983
+ */
7984
+ prompt: string;
7985
+ /**
7986
+ * Rig Task Id
7987
+ * @description Rigging task ID. Only present when enable_rigging is true.
7988
+ */
7989
+ rig_task_id?: string;
7990
+ rigged_character_fbx?: Components.File;
7991
+ rigged_character_glb?: Components.File;
7992
+ /**
7993
+ * Seed
7994
+ * @description The seed used for generation
7995
+ * @example 4002110719
7996
+ */
7997
+ seed?: number;
7998
+ /**
7999
+ * Texture Urls
8000
+ * @description Array of texture file objects
8001
+ * @example [
8002
+ * {
8003
+ * "base_color": {
8004
+ * "file_size": 4254502,
8005
+ * "file_name": "texture_0.png",
8006
+ * "content_type": "image/png",
8007
+ * "url": "https://v3b.fal.media/files/b/panda/DoPKAuZY0tTjnr6C9ee-Q_texture_0.png"
8008
+ * }
8009
+ * }
8010
+ * ]
8011
+ */
8012
+ texture_urls?: Components.TextureFiles[];
8013
+ /**
8014
+ * @description Preview thumbnail of the generated model
8015
+ * @example {
8016
+ * "file_size": 173792,
8017
+ * "file_name": "preview.png",
8018
+ * "content_type": "image/png",
8019
+ * "url": "https://v3b.fal.media/files/b/koala/6LJISu4ilkZXcdOETwl_d_preview.png"
8020
+ * }
8021
+ */
8022
+ thumbnail?: Components.File;
8023
+ }
8024
+
8020
8025
  export interface SharedType_751 {
8021
8026
  video: Components.File;
8022
8027
  }
@@ -9005,6 +9010,90 @@ export interface SharedType_5f1 {
9005
9010
  };
9006
9011
  }
9007
9012
 
9013
+ export interface SharedType_5ac {
9014
+ /**
9015
+ * Acoustic Cfg Scale
9016
+ * @description Classifier-free guidance scale for acoustic feature generation.
9017
+ * @default 1.6
9018
+ */
9019
+ acoustic_cfg_scale?: number;
9020
+ /**
9021
+ * Audio Url
9022
+ * @description URL of the reference audio file for voice cloning. The model will replicate this speaker's voice characteristics.
9023
+ * @example https://storage.googleapis.com/falserverless/example_inputs/reference_audio.wav
9024
+ */
9025
+ audio_url: string;
9026
+ /**
9027
+ * Language
9028
+ * @description Language for text alignment. Use the appropriate code for non-English synthesis.
9029
+ * @default en
9030
+ * @enum {string}
9031
+ */
9032
+ language?: 'en' | 'ar' | 'ch' | 'de' | 'es' | 'fr' | 'it' | 'ja' | 'pl' | 'pt';
9033
+ /**
9034
+ * Noise Temperature
9035
+ * @description Temperature for noise in the flow matching diffusion process.
9036
+ * @default 0.9
9037
+ */
9038
+ noise_temperature?: number;
9039
+ /**
9040
+ * Num Extra Steps
9041
+ * @description Number of extra autoregressive steps for speech continuation beyond the input text. Useful for generating trailing prosody or silence.
9042
+ * @default 0
9043
+ */
9044
+ num_extra_steps?: number;
9045
+ /**
9046
+ * Num Inference Steps
9047
+ * @description Number of ODE solver steps for flow matching acoustic generation. More steps improve quality at the cost of speed.
9048
+ * @default 20
9049
+ */
9050
+ num_inference_steps?: number;
9051
+ /**
9052
+ * Output Format
9053
+ * @description The format of the output audio file.
9054
+ * @default wav
9055
+ * @enum {string}
9056
+ */
9057
+ output_format?: 'wav' | 'mp3';
9058
+ /**
9059
+ * Prompt
9060
+ * @description The text to synthesize into speech using the reference speaker's voice.
9061
+ * @example The autumn leaves danced gracefully through the crisp morning air, painting the cobblestone streets with shades of amber and gold.
9062
+ */
9063
+ prompt: string;
9064
+ /**
9065
+ * Repetition Penalty
9066
+ * @description Penalty applied to repeated tokens during generation.
9067
+ * @default 1.1
9068
+ */
9069
+ repetition_penalty?: number;
9070
+ /**
9071
+ * Speed Up Factor
9072
+ * @description Factor to speed up or slow down the generated speech. Values > 1.0 speed up, < 1.0 slow down.
9073
+ * @default 1
9074
+ */
9075
+ speed_up_factor?: number;
9076
+ /**
9077
+ * Temperature
9078
+ * @description Sampling temperature for text token generation. Higher values produce more varied output.
9079
+ * @default 0.6
9080
+ */
9081
+ temperature?: number;
9082
+ /**
9083
+ * Top P
9084
+ * @description Top-p (nucleus) sampling parameter for text generation.
9085
+ * @default 0.9
9086
+ */
9087
+ top_p?: number;
9088
+ /**
9089
+ * Transcript
9090
+ * @description Transcript of the reference audio. For non-English audio, providing a transcript is required since the built-in ASR is English-only.
9091
+ * @default
9092
+ * @example Some call me nature, others call me mother nature.
9093
+ */
9094
+ transcript?: string;
9095
+ }
9096
+
9008
9097
  export interface SharedType_5a8 {
9009
9098
  /**
9010
9099
  * Acceleration
@@ -10006,6 +10095,105 @@ export interface SharedType_4d6 {
10006
10095
  };
10007
10096
  }
10008
10097
 
10098
+ export interface SharedType_4c9 {
10099
+ /**
10100
+ * Control Lora Image Url
10101
+ * @description The image to use for control lora. This is used to control the style of the generated image.
10102
+ * @example https://v3.fal.media/files/kangaroo/Cb7BeM7G4DauK_lWjzY3N_Celeb6.jpg
10103
+ */
10104
+ control_lora_image_url: string;
10105
+ /**
10106
+ * Control Lora Strength
10107
+ * @description The strength of the control lora.
10108
+ * @default 1
10109
+ */
10110
+ control_lora_strength?: number;
10111
+ /**
10112
+ * Enable Safety Checker
10113
+ * @description If set to true, the safety checker will be enabled.
10114
+ * @default true
10115
+ */
10116
+ enable_safety_checker?: boolean;
10117
+ /**
10118
+ * Guidance scale (CFG)
10119
+ * @description The CFG (Classifier Free Guidance) scale is a measure of how close you want
10120
+ * the model to stick to your prompt when looking for a related image to show you.
10121
+ * @default 3.5
10122
+ */
10123
+ guidance_scale?: number;
10124
+ /**
10125
+ * Image Size
10126
+ * @description The size of the generated image.
10127
+ * @default landscape_4_3
10128
+ */
10129
+ image_size?:
10130
+ | Components.ImageSize
10131
+ | (
10132
+ | 'square_hd'
10133
+ | 'square'
10134
+ | 'portrait_4_3'
10135
+ | 'portrait_16_9'
10136
+ | 'landscape_4_3'
10137
+ | 'landscape_16_9'
10138
+ );
10139
+ /**
10140
+ * Image Url
10141
+ * @description URL of image to use for image-to-image generation.
10142
+ */
10143
+ image_url?: string;
10144
+ /**
10145
+ * Loras
10146
+ * @description The LoRAs to use for the image generation. You can use any number of LoRAs
10147
+ * and they will be merged together to generate the final image.
10148
+ * @default []
10149
+ */
10150
+ loras?: Components.LoraWeight[];
10151
+ /**
10152
+ * Num Images
10153
+ * @description The number of images to generate.
10154
+ * @default 1
10155
+ */
10156
+ num_images?: number;
10157
+ /**
10158
+ * Num Inference Steps
10159
+ * @description The number of inference steps to perform.
10160
+ * @default 28
10161
+ */
10162
+ num_inference_steps?: number;
10163
+ /**
10164
+ * Output Format
10165
+ * @description The format of the generated image.
10166
+ * @default jpeg
10167
+ * @enum {string}
10168
+ */
10169
+ output_format?: 'jpeg' | 'png';
10170
+ /**
10171
+ * Preprocess Depth
10172
+ * @description If set to true, the input image will be preprocessed to extract depth information.
10173
+ * This is useful for generating depth maps from images.
10174
+ * @default true
10175
+ */
10176
+ preprocess_depth?: boolean;
10177
+ /**
10178
+ * Prompt
10179
+ * @description The prompt to generate an image from.
10180
+ * @example Extreme close-up of a single tiger eye, direct frontal view. Detailed iris and pupil. Sharp focus on eye texture and color. Natural lighting to capture authentic eye shine and depth. The word "FLUX" is painted over it in big, white brush strokes with visible texture.
10181
+ */
10182
+ prompt: string;
10183
+ /**
10184
+ * Seed
10185
+ * @description The same seed and the same prompt given to the same version of the model
10186
+ * will output the same image every time.
10187
+ */
10188
+ seed?: number;
10189
+ /**
10190
+ * Sync Mode
10191
+ * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
10192
+ * @default false
10193
+ */
10194
+ sync_mode?: boolean;
10195
+ }
10196
+
10009
10197
  export interface SharedType_4c2 {
10010
10198
  /**
10011
10199
  * Description
@@ -10242,6 +10430,99 @@ export interface SharedType_4a0 {
10242
10430
  sync_mode?: boolean;
10243
10431
  }
10244
10432
 
10433
+ export interface SharedType_49b {
10434
+ /**
10435
+ * Acceleration
10436
+ * @description Acceleration level for image generation. 'regular' balances speed and quality.
10437
+ * @default none
10438
+ * @enum {string}
10439
+ */
10440
+ acceleration?: 'none' | 'regular';
10441
+ /**
10442
+ * Enable Safety Checker
10443
+ * @description If set to true, the safety checker will be enabled.
10444
+ * @default true
10445
+ */
10446
+ enable_safety_checker?: boolean;
10447
+ /**
10448
+ * Guidance scale (CFG)
10449
+ * @description The CFG (Classifier Free Guidance) scale is a measure of how close you want
10450
+ * the model to stick to your prompt when looking for a related image to show you.
10451
+ * @default 3.5
10452
+ */
10453
+ guidance_scale?: number;
10454
+ /**
10455
+ * Image Size
10456
+ * @description The size of the generated image.
10457
+ */
10458
+ image_size?:
10459
+ | Components.ImageSize
10460
+ | (
10461
+ | 'square_hd'
10462
+ | 'square'
10463
+ | 'portrait_4_3'
10464
+ | 'portrait_16_9'
10465
+ | 'landscape_4_3'
10466
+ | 'landscape_16_9'
10467
+ );
10468
+ /**
10469
+ * Image Url
10470
+ * @description URL of image to use for inpainting. or img2img
10471
+ * @example https://storage.googleapis.com/falserverless/example_inputs/dog.png
10472
+ */
10473
+ image_url: string;
10474
+ /**
10475
+ * Loras
10476
+ * @description The LoRAs to use for the image generation. You can use any number of LoRAs
10477
+ * and they will be merged together to generate the final image.
10478
+ * @default []
10479
+ */
10480
+ loras?: Components.LoraWeight[];
10481
+ /**
10482
+ * Num Images
10483
+ * @description The number of images to generate. This is always set to 1 for streaming output.
10484
+ * @default 1
10485
+ */
10486
+ num_images?: number;
10487
+ /**
10488
+ * Num Inference Steps
10489
+ * @description The number of inference steps to perform.
10490
+ * @default 28
10491
+ */
10492
+ num_inference_steps?: number;
10493
+ /**
10494
+ * Output Format
10495
+ * @description The format of the generated image.
10496
+ * @default jpeg
10497
+ * @enum {string}
10498
+ */
10499
+ output_format?: 'jpeg' | 'png';
10500
+ /**
10501
+ * Prompt
10502
+ * @description The prompt to generate an image from.
10503
+ * @example A photo of a lion sitting on a stone bench
10504
+ */
10505
+ prompt: string;
10506
+ /**
10507
+ * Seed
10508
+ * @description The same seed and the same prompt given to the same version of the model
10509
+ * will output the same image every time.
10510
+ */
10511
+ seed?: number;
10512
+ /**
10513
+ * Strength
10514
+ * @description The strength to use for inpainting/image-to-image. Only used if the image_url is provided. 1.0 is completely remakes the image while 0.0 preserves the original.
10515
+ * @default 0.85
10516
+ */
10517
+ strength?: number;
10518
+ /**
10519
+ * Sync Mode
10520
+ * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
10521
+ * @default false
10522
+ */
10523
+ sync_mode?: boolean;
10524
+ }
10525
+
10245
10526
  export interface SharedType_484 {
10246
10527
  /**
10247
10528
  * Seed
@@ -12019,81 +12300,6 @@ export interface SharedType_20d {
12019
12300
  seed: number;
12020
12301
  }
12021
12302
 
12022
- export interface SharedType_207 {
12023
- /**
12024
- * Enable Safety Checker
12025
- * @description If set to true, the safety checker will be enabled.
12026
- * @default true
12027
- */
12028
- enable_safety_checker?: boolean;
12029
- /**
12030
- * Guidance scale (CFG)
12031
- * @description The CFG (Classifier Free Guidance) scale is a measure of how close you want
12032
- * the model to stick to your prompt when looking for a related image to show you.
12033
- * @default 3.5
12034
- */
12035
- guidance_scale?: number;
12036
- /**
12037
- * Image Size
12038
- * @description The size of the generated image.
12039
- * @default landscape_4_3
12040
- */
12041
- image_size?:
12042
- | Components.ImageSize
12043
- | (
12044
- | 'square_hd'
12045
- | 'square'
12046
- | 'portrait_4_3'
12047
- | 'portrait_16_9'
12048
- | 'landscape_4_3'
12049
- | 'landscape_16_9'
12050
- );
12051
- /**
12052
- * Loras
12053
- * @description The LoRAs to use for the image generation. You can use any number of LoRAs
12054
- * and they will be merged together to generate the final image.
12055
- * @default []
12056
- */
12057
- loras?: Components.LoraWeight[];
12058
- /**
12059
- * Num Images
12060
- * @description The number of images to generate. This is always set to 1 for streaming output.
12061
- * @default 1
12062
- */
12063
- num_images?: number;
12064
- /**
12065
- * Num Inference Steps
12066
- * @description The number of inference steps to perform.
12067
- * @default 28
12068
- */
12069
- num_inference_steps?: number;
12070
- /**
12071
- * Output Format
12072
- * @description The format of the generated image.
12073
- * @default jpeg
12074
- * @enum {string}
12075
- */
12076
- output_format?: 'jpeg' | 'png';
12077
- /**
12078
- * Prompt
12079
- * @description The prompt to generate an image from.
12080
- * @example Extreme close-up of a single tiger eye, direct frontal view. Detailed iris and pupil. Sharp focus on eye texture and color. Natural lighting to capture authentic eye shine and depth. The word "FLUX" is painted over it in big, white brush strokes with visible texture.
12081
- */
12082
- prompt: string;
12083
- /**
12084
- * Seed
12085
- * @description The same seed and the same prompt given to the same version of the model
12086
- * will output the same image every time.
12087
- */
12088
- seed?: number;
12089
- /**
12090
- * Sync Mode
12091
- * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
12092
- * @default false
12093
- */
12094
- sync_mode?: boolean;
12095
- }
12096
-
12097
12303
  export interface SharedType_1eb {
12098
12304
  /**
12099
12305
  * @description The generated video file.
@@ -16364,9 +16570,7 @@ export interface WorkflowUtilitiesAutoSubtitleOutput {
16364
16570
  * Transcription Metadata
16365
16571
  * @description Additional transcription metadata from ElevenLabs (language, segments, etc.)
16366
16572
  */
16367
- transcription_metadata?: {
16368
- [key: string]: { [x: string]: any } | null;
16369
- };
16573
+ transcription_metadata?: Record<string, number>;
16370
16574
  /**
16371
16575
  * @description The video with automatic subtitles
16372
16576
  * @example {
@@ -16381,9 +16585,7 @@ export interface WorkflowUtilitiesAutoSubtitleOutput {
16381
16585
  * Words
16382
16586
  * @description Word-level timing information from transcription service
16383
16587
  */
16384
- words?: {
16385
- [key: string]: { [x: string]: any } | null;
16386
- }[];
16588
+ words?: Record<string, never>[];
16387
16589
  }
16388
16590
 
16389
16591
  export interface WorkflowUtilitiesAudioCompressorInput {
@@ -24014,6 +24216,30 @@ export interface TheraInput {
24014
24216
 
24015
24217
  export interface TheraOutput extends SharedType_678 {}
24016
24218
 
24219
+ export interface Tada3bTextToSpeechInput extends SharedType_5ac {}
24220
+
24221
+ export interface Tada3bTextToSpeechOutput {
24222
+ /**
24223
+ * @description The generated speech audio file.
24224
+ * @example {
24225
+ * "url": "https://v3b.fal.media/files/b/0a91e54b/Ldhq_0EWjUdnUEwn-Tcrk_nfs9DTGI.wav"
24226
+ * }
24227
+ */
24228
+ audio: Components.AudioFile;
24229
+ }
24230
+
24231
+ export interface Tada1bTextToSpeechInput extends SharedType_5ac {}
24232
+
24233
+ export interface Tada1bTextToSpeechOutput {
24234
+ /**
24235
+ * @description The generated speech audio file.
24236
+ * @example {
24237
+ * "url": "https://v3b.fal.media/files/b/0a91e55e/HUVJsjW6QoDUBfg6HdM5e_B2R8G4X4.wav"
24238
+ * }
24239
+ */
24240
+ audio: Components.AudioFile;
24241
+ }
24242
+
24017
24243
  export interface T2vTurboInput {
24018
24244
  /**
24019
24245
  * Export Fps
@@ -24177,11 +24403,11 @@ export interface SyncLipsyncReact1Output {
24177
24403
  * "height": 1088,
24178
24404
  * "duration": 7.041667,
24179
24405
  * "url": "https://storage.googleapis.com/falserverless/example_outputs/react_1/output.mp4",
24180
- * "width": 1920,
24181
24406
  * "fps": 24,
24407
+ * "width": 1920,
24182
24408
  * "file_name": "output.mp4",
24183
- * "num_frames": 169,
24184
- * "content_type": "video/mp4"
24409
+ * "content_type": "video/mp4",
24410
+ * "num_frames": 169
24185
24411
  * }
24186
24412
  */
24187
24413
  video: Components.VideoFile;
@@ -25654,6 +25880,11 @@ export interface Sora2TextToVideoProInput {
25654
25880
  * @enum {string}
25655
25881
  */
25656
25882
  aspect_ratio?: '9:16' | '16:9';
25883
+ /**
25884
+ * Character IDs
25885
+ * @description Up to two character IDs (from create-character) to use in the video. Refer to characters by name in the prompt. When set, only the OpenAI provider is used.
25886
+ */
25887
+ character_ids?: string[];
25657
25888
  /**
25658
25889
  * Delete Video
25659
25890
  * @description Whether to delete the video after generation for privacy reasons. If True, the video cannot be used for remixing and will be permanently deleted.
@@ -25672,7 +25903,7 @@ export interface Sora2TextToVideoProInput {
25672
25903
  * @default 4
25673
25904
  * @enum {integer}
25674
25905
  */
25675
- duration?: 4 | 8 | 12;
25906
+ duration?: 4 | 8 | 12 | 16 | 20;
25676
25907
  /**
25677
25908
  * Prompt
25678
25909
  * @description The text prompt describing the video you want to generate
@@ -25685,7 +25916,7 @@ export interface Sora2TextToVideoProInput {
25685
25916
  * @default 1080p
25686
25917
  * @enum {string}
25687
25918
  */
25688
- resolution?: '720p' | '1080p';
25919
+ resolution?: '720p' | '1080p' | 'true_1080p';
25689
25920
  }
25690
25921
 
25691
25922
  export interface Sora2TextToVideoProOutput {
@@ -25715,6 +25946,11 @@ export interface Sora2TextToVideoInput {
25715
25946
  * @enum {string}
25716
25947
  */
25717
25948
  aspect_ratio?: '9:16' | '16:9';
25949
+ /**
25950
+ * Character IDs
25951
+ * @description Up to two character IDs (from create-character) to use in the video. Refer to characters by name in the prompt. When set, only the OpenAI provider is used.
25952
+ */
25953
+ character_ids?: string[];
25718
25954
  /**
25719
25955
  * Delete Video
25720
25956
  * @description Whether to delete the video after generation for privacy reasons. If True, the video cannot be used for remixing and will be permanently deleted.
@@ -25733,7 +25969,7 @@ export interface Sora2TextToVideoInput {
25733
25969
  * @default 4
25734
25970
  * @enum {integer}
25735
25971
  */
25736
- duration?: 4 | 8 | 12;
25972
+ duration?: 4 | 8 | 12 | 16 | 20;
25737
25973
  /**
25738
25974
  * Model
25739
25975
  * @description The model to use for the generation. When the default model is selected, the latest snapshot of the model will be used - otherwise, select a specific snapshot of the model.
@@ -25783,6 +26019,11 @@ export interface Sora2ImageToVideoProInput {
25783
26019
  * @enum {string}
25784
26020
  */
25785
26021
  aspect_ratio?: 'auto' | '9:16' | '16:9';
26022
+ /**
26023
+ * Character IDs
26024
+ * @description Up to two character IDs (from create-character) to use in the video. Refer to characters by name in the prompt. When set, only the OpenAI provider is used.
26025
+ */
26026
+ character_ids?: string[];
25786
26027
  /**
25787
26028
  * Delete Video
25788
26029
  * @description Whether to delete the video after generation for privacy reasons. If True, the video cannot be used for remixing and will be permanently deleted.
@@ -25801,7 +26042,7 @@ export interface Sora2ImageToVideoProInput {
25801
26042
  * @default 4
25802
26043
  * @enum {integer}
25803
26044
  */
25804
- duration?: 4 | 8 | 12;
26045
+ duration?: 4 | 8 | 12 | 16 | 20;
25805
26046
  /**
25806
26047
  * Image URL
25807
26048
  * @description The URL of the image to use as the first frame
@@ -25820,7 +26061,7 @@ export interface Sora2ImageToVideoProInput {
25820
26061
  * @default auto
25821
26062
  * @enum {string}
25822
26063
  */
25823
- resolution?: 'auto' | '720p' | '1080p';
26064
+ resolution?: 'auto' | '720p' | '1080p' | 'true_1080p';
25824
26065
  }
25825
26066
 
25826
26067
  export interface Sora2ImageToVideoProOutput {
@@ -25850,6 +26091,11 @@ export interface Sora2ImageToVideoInput {
25850
26091
  * @enum {string}
25851
26092
  */
25852
26093
  aspect_ratio?: 'auto' | '9:16' | '16:9';
26094
+ /**
26095
+ * Character IDs
26096
+ * @description Up to two character IDs (from create-character) to use in the video. Refer to characters by name in the prompt. When set, only the OpenAI provider is used.
26097
+ */
26098
+ character_ids?: string[];
25853
26099
  /**
25854
26100
  * Delete Video
25855
26101
  * @description Whether to delete the video after generation for privacy reasons. If True, the video cannot be used for remixing and will be permanently deleted.
@@ -25868,7 +26114,7 @@ export interface Sora2ImageToVideoInput {
25868
26114
  * @default 4
25869
26115
  * @enum {integer}
25870
26116
  */
25871
- duration?: 4 | 8 | 12;
26117
+ duration?: 4 | 8 | 12 | 16 | 20;
25872
26118
  /**
25873
26119
  * Image URL
25874
26120
  * @description The URL of the image to use as the first frame
@@ -25916,6 +26162,33 @@ export interface Sora2ImageToVideoOutput {
25916
26162
  video_id: string;
25917
26163
  }
25918
26164
 
26165
+ export interface Sora2CharactersInput {
26166
+ /**
26167
+ * Name
26168
+ * @description Name for the character (1–80 characters). Refer to this name in prompts when using the character.
26169
+ */
26170
+ name: string;
26171
+ /**
26172
+ * Video URL
26173
+ * @description URL of an MP4 video (minimum 720p, max ~2.67:1 aspect ratio) to define the character. Videos exceeding 1080p are automatically scaled down. Non-standard aspect ratios are automatically padded to 16:9 (landscape) or 9:16 (portrait). Videos longer than 4 seconds are trimmed to the first 4 seconds.
26174
+ */
26175
+ video_url: string;
26176
+ }
26177
+
26178
+ export interface Sora2CharactersOutput {
26179
+ /**
26180
+ * Character ID
26181
+ * @description API character ID (format char_...). Use this in character_ids when generating video.
26182
+ * @example char_123
26183
+ */
26184
+ id: string;
26185
+ /**
26186
+ * Name
26187
+ * @description The character name
26188
+ */
26189
+ name: string;
26190
+ }
26191
+
25919
26192
  export interface SmartTurnInput {
25920
26193
  /**
25921
26194
  * Audio Url
@@ -31874,7 +32147,6 @@ export interface PshumanInput {
31874
32147
 
31875
32148
  export interface PshumanOutput {
31876
32149
  /**
31877
- * Model Obj
31878
32150
  * @description The generated 3D model in OBJ format.
31879
32151
  * @example {
31880
32152
  * "file_name": "VGSdkXIgccoKhHs_JtXTa_result_clr_scale4_image.obj",
@@ -31882,9 +32154,8 @@ export interface PshumanOutput {
31882
32154
  * "url": "https://storage.googleapis.com/falserverless/model_tests/video_models/VGSdkXIgccoKhHs_JtXTa_result_clr_scale4_image.obj"
31883
32155
  * }
31884
32156
  */
31885
- model_obj: Components.File_1;
32157
+ model_obj: Components.File;
31886
32158
  /**
31887
- * Preview Image
31888
32159
  * @description A preview image showing the input and the generated multi-view outputs.
31889
32160
  * @example {
31890
32161
  * "file_name": "WCN_SkT2-RwsGHlxCVHyn_image_preview.png",
@@ -31892,7 +32163,7 @@ export interface PshumanOutput {
31892
32163
  * "url": "https://storage.googleapis.com/falserverless/model_tests/video_models/WCN_SkT2-RwsGHlxCVHyn_image_preview.png"
31893
32164
  * }
31894
32165
  */
31895
- preview_image: Components.File_1;
32166
+ preview_image: Components.File;
31896
32167
  }
31897
32168
 
31898
32169
  export interface PostProcessingVignetteInput {
@@ -35065,6 +35336,98 @@ export interface PiflowOutput {
35065
35336
  seed: number;
35066
35337
  }
35067
35338
 
35339
+ export interface PhysicEditInput {
35340
+ /**
35341
+ * Enable Safety Checker
35342
+ * @description Enable content safety checking on the output.
35343
+ * @default true
35344
+ */
35345
+ enable_safety_checker?: boolean;
35346
+ /**
35347
+ * Guidance Scale
35348
+ * @description Classifier-free guidance scale. Higher values follow the prompt more closely.
35349
+ * @default 4
35350
+ */
35351
+ guidance_scale?: number;
35352
+ /**
35353
+ * Image URL
35354
+ * @description URL of the image to edit with physics-aware transformations.
35355
+ * @example https://v3b.fal.media/files/b/0a91d47e/NG382xuDf06EcPCaS2eW3_Signature_Ice_Sculptures-03.jpg
35356
+ */
35357
+ image_url: string;
35358
+ /**
35359
+ * Negative Prompt
35360
+ * @description Description of unwanted elements in the output.
35361
+ * @default
35362
+ */
35363
+ negative_prompt?: string;
35364
+ /**
35365
+ * Num Inference Steps
35366
+ * @description Number of denoising steps. More steps generally produce higher quality but take longer.
35367
+ * @default 40
35368
+ */
35369
+ num_inference_steps?: number;
35370
+ /**
35371
+ * Output Format
35372
+ * @description The format of the generated image.
35373
+ * @default jpeg
35374
+ * @enum {string}
35375
+ */
35376
+ output_format?: 'jpeg' | 'png';
35377
+ /**
35378
+ * Prompt
35379
+ * @description The editing instruction describing the physical transformation to apply.
35380
+ * @example Make the ice sculpture melt gradually, with water pooling beneath and light refracting through the translucent surface
35381
+ */
35382
+ prompt: string;
35383
+ /**
35384
+ * Seed
35385
+ * @description Random seed for reproducibility. Same seed and prompt produce identical output.
35386
+ */
35387
+ seed?: number;
35388
+ /**
35389
+ * Sync Mode
35390
+ * @description If True, the image will be returned as a data URI.
35391
+ * @default false
35392
+ */
35393
+ sync_mode?: boolean;
35394
+ }
35395
+
35396
+ export interface PhysicEditOutput {
35397
+ /**
35398
+ * Has Nsfw Concepts
35399
+ * @description Whether the generated images contain NSFW concepts.
35400
+ */
35401
+ has_nsfw_concepts: boolean[];
35402
+ /**
35403
+ * Images
35404
+ * @description The edited images.
35405
+ * @example [
35406
+ * {
35407
+ * "height": 1152,
35408
+ * "content_type": "image/jpeg",
35409
+ * "url": "https://v3b.fal.media/files/b/0a91cf62/DdpKY15kkD6N6v66uJ0q1.jpg",
35410
+ * "width": 864
35411
+ * }
35412
+ * ]
35413
+ */
35414
+ images: Components.Image[];
35415
+ /**
35416
+ * Prompt
35417
+ * @description The prompt used for generating the image.
35418
+ */
35419
+ prompt: string;
35420
+ /**
35421
+ * Seed
35422
+ * @description Seed of the generated Image. It will be the same value of the one passed in the
35423
+ * input or the randomly generated that was used in case none was passed.
35424
+ */
35425
+ seed: number;
35426
+ timings: {
35427
+ [key: string]: number;
35428
+ };
35429
+ }
35430
+
35068
35431
  export interface PhotomakerInput {
35069
35432
  /**
35070
35433
  * Base Pipeline
@@ -36344,13 +36707,13 @@ export interface NovaSrOutput {
36344
36707
  /**
36345
36708
  * @description The enhanced audio file.
36346
36709
  * @example {
36347
- * "channels": 1,
36710
+ * "bitrate": "192k",
36348
36711
  * "duration": 12.283291666666667,
36349
36712
  * "url": "https://v3b.fal.media/files/b/0a8a3f1a/lTKExJu-R6ZJdnFlpzEeq_TxmNTNhl.mp3",
36350
36713
  * "file_name": "lTKExJu-R6ZJdnFlpzEeq_TxmNTNhl.mp3",
36351
36714
  * "sample_rate": 48000,
36352
36715
  * "content_type": "audio/mpeg",
36353
- * "bitrate": "192k"
36716
+ * "channels": 1
36354
36717
  * }
36355
36718
  */
36356
36719
  audio: Components.AudioFile;
@@ -36516,13 +36879,7 @@ export interface MusetalkInput {
36516
36879
  source_video_url: string;
36517
36880
  }
36518
36881
 
36519
- export interface MusetalkOutput {
36520
- /**
36521
- * Video
36522
- * @description The generated video file.
36523
- */
36524
- video: Components.File_1;
36525
- }
36882
+ export interface MusetalkOutput extends SharedType_751 {}
36526
36883
 
36527
36884
  export interface MultishotMasterInput {
36528
36885
  /**
@@ -38432,19 +38789,19 @@ export interface MinimaxMusicOutput {
38432
38789
 
38433
38790
  export interface MeshyV6TextTo3dInput extends SharedType_477 {}
38434
38791
 
38435
- export interface MeshyV6TextTo3dOutput extends SharedType_ba1 {}
38792
+ export interface MeshyV6TextTo3dOutput extends SharedType_752 {}
38436
38793
 
38437
38794
  export interface MeshyV6ImageTo3dInput extends SharedType_7ca {}
38438
38795
 
38439
- export interface MeshyV6ImageTo3dOutput extends SharedType_82d {}
38796
+ export interface MeshyV6ImageTo3dOutput extends SharedType_d53 {}
38440
38797
 
38441
38798
  export interface MeshyV6PreviewTextTo3dInput extends SharedType_477 {}
38442
38799
 
38443
- export interface MeshyV6PreviewTextTo3dOutput extends SharedType_ba1 {}
38800
+ export interface MeshyV6PreviewTextTo3dOutput extends SharedType_752 {}
38444
38801
 
38445
38802
  export interface MeshyV6PreviewImageTo3dInput extends SharedType_7ca {}
38446
38803
 
38447
- export interface MeshyV6PreviewImageTo3dOutput extends SharedType_82d {}
38804
+ export interface MeshyV6PreviewImageTo3dOutput extends SharedType_d53 {}
38448
38805
 
38449
38806
  export interface MeshyV5RetextureInput {
38450
38807
  /**
@@ -38515,17 +38872,17 @@ export interface MeshyV5RetextureOutput {
38515
38872
  * "content_type": "model/vnd.usdz+zip",
38516
38873
  * "url": "https://v3b.fal.media/files/b/panda/4ItUhLHiH4foEw30qcWZv_model.usdz"
38517
38874
  * },
38518
- * "glb": {
38519
- * "file_size": 4097640,
38520
- * "file_name": "model.glb",
38521
- * "content_type": "model/gltf-binary",
38522
- * "url": "https://v3b.fal.media/files/b/tiger/pU0TtsRTxXM6VnKEYTHSV_model.glb"
38523
- * },
38524
38875
  * "obj": {
38525
38876
  * "file_size": 2964508,
38526
38877
  * "file_name": "model.obj",
38527
38878
  * "content_type": "text/plain",
38528
38879
  * "url": "https://v3b.fal.media/files/b/zebra/M5aK_b6vKH7KeGCZoSLq7_model.obj"
38880
+ * },
38881
+ * "glb": {
38882
+ * "file_size": 4097640,
38883
+ * "file_name": "model.glb",
38884
+ * "content_type": "model/gltf-binary",
38885
+ * "url": "https://v3b.fal.media/files/b/tiger/pU0TtsRTxXM6VnKEYTHSV_model.glb"
38529
38886
  * }
38530
38887
  * }
38531
38888
  */
@@ -38776,17 +39133,17 @@ export interface MeshyV5MultiImageTo3dOutput {
38776
39133
  * "content_type": "model/vnd.usdz+zip",
38777
39134
  * "url": "https://v3b.fal.media/files/b/panda/fSGLGmtgzUjhepklN06Zw_model.usdz"
38778
39135
  * },
38779
- * "glb": {
38780
- * "file_size": 7875308,
38781
- * "file_name": "model.glb",
38782
- * "content_type": "model/gltf-binary",
38783
- * "url": "https://v3b.fal.media/files/b/tiger/62QMEQqZ3pjUds4DfuVtX_model.glb"
38784
- * },
38785
39136
  * "obj": {
38786
39137
  * "file_size": 2761323,
38787
39138
  * "file_name": "model.obj",
38788
39139
  * "content_type": "text/plain",
38789
39140
  * "url": "https://v3b.fal.media/files/b/koala/xmOnmSeePfuROe3pqHpf0_model.obj"
39141
+ * },
39142
+ * "glb": {
39143
+ * "file_size": 7875308,
39144
+ * "file_name": "model.glb",
39145
+ * "content_type": "model/gltf-binary",
39146
+ * "url": "https://v3b.fal.media/files/b/tiger/62QMEQqZ3pjUds4DfuVtX_model.glb"
38790
39147
  * }
38791
39148
  * }
38792
39149
  */
@@ -39514,6 +39871,52 @@ export interface Lyria2Output {
39514
39871
  audio: Components.File;
39515
39872
  }
39516
39873
 
39874
+ export interface LuxTtsInput {
39875
+ /**
39876
+ * Audio Url
39877
+ * @description URL of the reference audio file for voice cloning. The model will mimic the voice characteristics from this audio.
39878
+ * @example https://storage.googleapis.com/falserverless/example_inputs/reference_audio.wav
39879
+ */
39880
+ audio_url: string;
39881
+ /**
39882
+ * Guidance Scale
39883
+ * @description Classifier-free guidance scale. Higher values increase adherence to the reference voice at the cost of diversity.
39884
+ * @default 3
39885
+ */
39886
+ guidance_scale?: number;
39887
+ /**
39888
+ * Max Ref Length
39889
+ * @description Maximum length of the reference audio to use for voice encoding, in seconds. Longer durations capture more voice characteristics but increase processing time.
39890
+ * @default 5
39891
+ */
39892
+ max_ref_length?: number;
39893
+ /**
39894
+ * Num Inference Steps
39895
+ * @description Number of flow-matching inference steps. 4 is recommended for best efficiency.
39896
+ * @default 4
39897
+ */
39898
+ num_inference_steps?: number;
39899
+ /**
39900
+ * Prompt
39901
+ * @description The text to be converted to speech.
39902
+ * @example Hey, what's up? I'm feeling really great today! The sun is shining and there's a gentle breeze rustling through the trees.
39903
+ */
39904
+ prompt: string;
39905
+ /**
39906
+ * Seed
39907
+ * @description Random seed for reproducibility.
39908
+ */
39909
+ seed?: number;
39910
+ }
39911
+
39912
+ export interface LuxTtsOutput {
39913
+ audio: Components.File;
39914
+ seed: number;
39915
+ timings: {
39916
+ [key: string]: number;
39917
+ };
39918
+ }
39919
+
39517
39920
  export interface LuminaImageV2Input {
39518
39921
  /**
39519
39922
  * Cfg Normalization
@@ -41831,7 +42234,7 @@ export interface Ltx2ImageToVideoOutput extends SharedType_62e {}
41831
42234
  export interface Ltx2ExtendVideoInput {
41832
42235
  /**
41833
42236
  * Context
41834
- * @description Number of seconds from the input video to use as context for the extension (maximum 20 seconds). If not provided, defaults to maximize available context within the 505 frame limit.
42237
+ * @description Number of seconds from the input video to use as context for the extension (minimum 1 second, maximum 20 seconds). If not provided, defaults to maximize available context within the 505 frame limit.
41835
42238
  */
41836
42239
  context?: number;
41837
42240
  /**
@@ -42202,7 +42605,7 @@ export interface Ltx23ImageToVideoOutput {
42202
42605
  export interface Ltx23ExtendVideoInput {
42203
42606
  /**
42204
42607
  * Context
42205
- * @description Number of seconds from the input video to use as context for the extension (maximum 20 seconds). If not provided, defaults to maximize available context within the 505 frame limit.
42608
+ * @description Number of seconds from the input video to use as context for the extension (minimum 1 second, maximum 20 seconds). If not provided, defaults to maximize available context within the 505 frame limit.
42206
42609
  */
42207
42610
  context?: number;
42208
42611
  /**
@@ -49993,13 +50396,6 @@ export interface KlingVideoV3StandardMotionControlOutput {
49993
50396
  }
49994
50397
 
49995
50398
  export interface KlingVideoV3StandardImageToVideoInput {
49996
- /**
49997
- * Aspect Ratio
49998
- * @description The aspect ratio of the generated video frame
49999
- * @default 16:9
50000
- * @enum {string}
50001
- */
50002
- aspect_ratio?: '16:9' | '9:16' | '1:1';
50003
50399
  /**
50004
50400
  * Cfg Scale
50005
50401
  * @description The CFG (Classifier Free Guidance) scale is a measure of how close you want
@@ -50214,13 +50610,6 @@ export interface KlingVideoV3ProMotionControlOutput {
50214
50610
  }
50215
50611
 
50216
50612
  export interface KlingVideoV3ProImageToVideoInput {
50217
- /**
50218
- * Aspect Ratio
50219
- * @description The aspect ratio of the generated video frame
50220
- * @default 16:9
50221
- * @enum {string}
50222
- */
50223
- aspect_ratio?: '16:9' | '9:16' | '1:1';
50224
50613
  /**
50225
50614
  * Cfg Scale
50226
50615
  * @description The CFG (Classifier Free Guidance) scale is a measure of how close you want
@@ -59552,8 +59941,8 @@ export interface Hunyuan_worldOutput {
59552
59941
  /**
59553
59942
  * @description The generated panorama image.
59554
59943
  * @example {
59555
- * "file_size": 2738127,
59556
59944
  * "height": 960,
59945
+ * "file_size": 2738127,
59557
59946
  * "file_name": "5db7925423b44f2a98098cd8f7cad7ec.png",
59558
59947
  * "content_type": "image/png",
59559
59948
  * "url": "https://v3.fal.media/files/kangaroo/P2AmXuLlyDIsivqjV_rAr_5db7925423b44f2a98098cd8f7cad7ec.png",
@@ -63936,9 +64325,7 @@ export interface FluxPulidInput {
63936
64325
  start_step?: number;
63937
64326
  /**
63938
64327
  * Sync Mode
63939
- * @description If set to true, the function will wait for the image to be generated and uploaded
63940
- * before returning the response. This will increase the latency of the function but
63941
- * it allows you to get the image directly in the response without going through the CDN.
64328
+ * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
63942
64329
  * @default false
63943
64330
  */
63944
64331
  sync_mode?: boolean;
@@ -64496,7 +64883,81 @@ export interface FluxLoraStreamInput extends SharedType_c5e {}
64496
64883
 
64497
64884
  export interface FluxLoraStreamOutput extends SharedType_7c6 {}
64498
64885
 
64499
- export interface FluxLoraInpaintingInput {
64886
+ export interface FluxLoraInpaintingInput extends SharedType_90c {}
64887
+
64888
+ export interface FluxLoraInpaintingOutput extends SharedType_7c6 {}
64889
+
64890
+ export interface FluxLoraImageToImageInput extends SharedType_49b {}
64891
+
64892
+ export interface FluxLoraImageToImageOutput extends SharedType_7c6 {}
64893
+
64894
+ export interface FluxLoraPortraitTrainerInput {
64895
+ /**
64896
+ * Create Masks
64897
+ * @description If True, masks will be created for the subject.
64898
+ * @default false
64899
+ * @example false
64900
+ */
64901
+ create_masks?: boolean;
64902
+ /**
64903
+ * Data Archive Format
64904
+ * @description The format of the archive. If not specified, the format will be inferred from the URL.
64905
+ */
64906
+ data_archive_format?: string;
64907
+ /**
64908
+ * Images Data Url
64909
+ * @description URL to zip archive with images of a consistent style. Try to use at least 10 images, although more is better.
64910
+ *
64911
+ * In addition to images the archive can contain text files with captions. Each text file should have the same name as the image file it corresponds to.
64912
+ *
64913
+ * The captions can include a special string `[trigger]`. If a trigger_word is specified, it will replace `[trigger]` in the captions.
64914
+ */
64915
+ images_data_url: string;
64916
+ /**
64917
+ * Learning Rate
64918
+ * @description Learning rate to use for training.
64919
+ * @default 0.00009
64920
+ * @example 0.0002
64921
+ */
64922
+ learning_rate?: number;
64923
+ /**
64924
+ * Multiresolution Training
64925
+ * @description If True, multiresolution training will be used.
64926
+ * @default true
64927
+ * @example true
64928
+ */
64929
+ multiresolution_training?: boolean;
64930
+ /**
64931
+ * Resume From Checkpoint
64932
+ * @description URL to a checkpoint to resume training from.
64933
+ * @default
64934
+ */
64935
+ resume_from_checkpoint?: string;
64936
+ /**
64937
+ * Steps
64938
+ * @description Number of steps to train the LoRA on.
64939
+ * @default 2500
64940
+ * @example 1000
64941
+ */
64942
+ steps?: number;
64943
+ /**
64944
+ * Subject Crop
64945
+ * @description If True, the subject will be cropped from the image.
64946
+ * @default true
64947
+ * @example true
64948
+ */
64949
+ subject_crop?: boolean;
64950
+ /**
64951
+ * Trigger Phrase
64952
+ * @description Trigger phrase to be used in the captions. If None, a trigger word will not be used.
64953
+ * If no captions are provide the trigger_work will be used instead of captions. If captions are provided, the trigger word will replace the `[trigger]` string in the captions.
64954
+ */
64955
+ trigger_phrase?: string;
64956
+ }
64957
+
64958
+ export interface FluxLoraPortraitTrainerOutput extends SharedType_9a6 {}
64959
+
64960
+ export interface FluxLoraFillInput {
64500
64961
  /**
64501
64962
  * Acceleration
64502
64963
  * @description Acceleration level for image generation. 'regular' balances speed and quality.
@@ -64510,11 +64971,12 @@ export interface FluxLoraInpaintingInput {
64510
64971
  * @default true
64511
64972
  */
64512
64973
  enable_safety_checker?: boolean;
64974
+ fill_image?: Components.ImageFillInput_1;
64513
64975
  /**
64514
64976
  * Guidance scale (CFG)
64515
64977
  * @description The CFG (Classifier Free Guidance) scale is a measure of how close you want
64516
64978
  * the model to stick to your prompt when looking for a related image to show you.
64517
- * @default 3.5
64979
+ * @default 30
64518
64980
  */
64519
64981
  guidance_scale?: number;
64520
64982
  /**
@@ -64533,8 +64995,8 @@ export interface FluxLoraInpaintingInput {
64533
64995
  );
64534
64996
  /**
64535
64997
  * Image Url
64536
- * @description URL of image to use for inpainting. or img2img
64537
- * @example https://storage.googleapis.com/falserverless/example_inputs/dog.png
64998
+ * @description URL of image to use for fill operation
64999
+ * @example https://storage.googleapis.com/falserverless/flux-lora/example-images/knight.jpeg
64538
65000
  */
64539
65001
  image_url: string;
64540
65002
  /**
@@ -64547,270 +65009,7 @@ export interface FluxLoraInpaintingInput {
64547
65009
  /**
64548
65010
  * Mask Url
64549
65011
  * @description The mask to area to Inpaint in.
64550
- * @example https://storage.googleapis.com/falserverless/example_inputs/dog_mask.png
64551
- */
64552
- mask_url: string;
64553
- /**
64554
- * Num Images
64555
- * @description The number of images to generate. This is always set to 1 for streaming output.
64556
- * @default 1
64557
- */
64558
- num_images?: number;
64559
- /**
64560
- * Num Inference Steps
64561
- * @description The number of inference steps to perform.
64562
- * @default 28
64563
- */
64564
- num_inference_steps?: number;
64565
- /**
64566
- * Output Format
64567
- * @description The format of the generated image.
64568
- * @default jpeg
64569
- * @enum {string}
64570
- */
64571
- output_format?: 'jpeg' | 'png';
64572
- /**
64573
- * Prompt
64574
- * @description The prompt to generate an image from.
64575
- * @example A photo of a lion sitting on a stone bench
64576
- */
64577
- prompt: string;
64578
- /**
64579
- * Seed
64580
- * @description The same seed and the same prompt given to the same version of the model
64581
- * will output the same image every time.
64582
- */
64583
- seed?: number;
64584
- /**
64585
- * Strength
64586
- * @description The strength to use for inpainting/image-to-image. Only used if the image_url is provided. 1.0 is completely remakes the image while 0.0 preserves the original.
64587
- * @default 0.85
64588
- */
64589
- strength?: number;
64590
- /**
64591
- * Sync Mode
64592
- * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
64593
- * @default false
64594
- */
64595
- sync_mode?: boolean;
64596
- }
64597
-
64598
- export interface FluxLoraInpaintingOutput extends SharedType_7c6 {}
64599
-
64600
- export interface FluxLoraImageToImageInput {
64601
- /**
64602
- * Acceleration
64603
- * @description Acceleration level for image generation. 'regular' balances speed and quality.
64604
- * @default none
64605
- * @enum {string}
64606
- */
64607
- acceleration?: 'none' | 'regular';
64608
- /**
64609
- * Enable Safety Checker
64610
- * @description If set to true, the safety checker will be enabled.
64611
- * @default true
64612
- */
64613
- enable_safety_checker?: boolean;
64614
- /**
64615
- * Guidance scale (CFG)
64616
- * @description The CFG (Classifier Free Guidance) scale is a measure of how close you want
64617
- * the model to stick to your prompt when looking for a related image to show you.
64618
- * @default 3.5
64619
- */
64620
- guidance_scale?: number;
64621
- /**
64622
- * Image Size
64623
- * @description The size of the generated image.
64624
- */
64625
- image_size?:
64626
- | Components.ImageSize
64627
- | (
64628
- | 'square_hd'
64629
- | 'square'
64630
- | 'portrait_4_3'
64631
- | 'portrait_16_9'
64632
- | 'landscape_4_3'
64633
- | 'landscape_16_9'
64634
- );
64635
- /**
64636
- * Image Url
64637
- * @description URL of image to use for inpainting. or img2img
64638
- * @example https://storage.googleapis.com/falserverless/example_inputs/dog.png
64639
- */
64640
- image_url: string;
64641
- /**
64642
- * Loras
64643
- * @description The LoRAs to use for the image generation. You can use any number of LoRAs
64644
- * and they will be merged together to generate the final image.
64645
- * @default []
64646
- */
64647
- loras?: Components.LoraWeight[];
64648
- /**
64649
- * Num Images
64650
- * @description The number of images to generate. This is always set to 1 for streaming output.
64651
- * @default 1
64652
- */
64653
- num_images?: number;
64654
- /**
64655
- * Num Inference Steps
64656
- * @description The number of inference steps to perform.
64657
- * @default 28
64658
- */
64659
- num_inference_steps?: number;
64660
- /**
64661
- * Output Format
64662
- * @description The format of the generated image.
64663
- * @default jpeg
64664
- * @enum {string}
64665
- */
64666
- output_format?: 'jpeg' | 'png';
64667
- /**
64668
- * Prompt
64669
- * @description The prompt to generate an image from.
64670
- * @example A photo of a lion sitting on a stone bench
64671
- */
64672
- prompt: string;
64673
- /**
64674
- * Seed
64675
- * @description The same seed and the same prompt given to the same version of the model
64676
- * will output the same image every time.
64677
- */
64678
- seed?: number;
64679
- /**
64680
- * Strength
64681
- * @description The strength to use for inpainting/image-to-image. Only used if the image_url is provided. 1.0 is completely remakes the image while 0.0 preserves the original.
64682
- * @default 0.85
64683
- */
64684
- strength?: number;
64685
- /**
64686
- * Sync Mode
64687
- * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
64688
- * @default false
64689
- */
64690
- sync_mode?: boolean;
64691
- }
64692
-
64693
- export interface FluxLoraImageToImageOutput extends SharedType_7c6 {}
64694
-
64695
- export interface FluxLoraPortraitTrainerInput {
64696
- /**
64697
- * Create Masks
64698
- * @description If True, masks will be created for the subject.
64699
- * @default false
64700
- * @example false
64701
- */
64702
- create_masks?: boolean;
64703
- /**
64704
- * Data Archive Format
64705
- * @description The format of the archive. If not specified, the format will be inferred from the URL.
64706
- */
64707
- data_archive_format?: string;
64708
- /**
64709
- * Images Data Url
64710
- * @description URL to zip archive with images of a consistent style. Try to use at least 10 images, although more is better.
64711
- *
64712
- * In addition to images the archive can contain text files with captions. Each text file should have the same name as the image file it corresponds to.
64713
- *
64714
- * The captions can include a special string `[trigger]`. If a trigger_word is specified, it will replace `[trigger]` in the captions.
64715
- */
64716
- images_data_url: string;
64717
- /**
64718
- * Learning Rate
64719
- * @description Learning rate to use for training.
64720
- * @default 0.00009
64721
- * @example 0.0002
64722
- */
64723
- learning_rate?: number;
64724
- /**
64725
- * Multiresolution Training
64726
- * @description If True, multiresolution training will be used.
64727
- * @default true
64728
- * @example true
64729
- */
64730
- multiresolution_training?: boolean;
64731
- /**
64732
- * Resume From Checkpoint
64733
- * @description URL to a checkpoint to resume training from.
64734
- * @default
64735
- */
64736
- resume_from_checkpoint?: string;
64737
- /**
64738
- * Steps
64739
- * @description Number of steps to train the LoRA on.
64740
- * @default 2500
64741
- * @example 1000
64742
- */
64743
- steps?: number;
64744
- /**
64745
- * Subject Crop
64746
- * @description If True, the subject will be cropped from the image.
64747
- * @default true
64748
- * @example true
64749
- */
64750
- subject_crop?: boolean;
64751
- /**
64752
- * Trigger Phrase
64753
- * @description Trigger phrase to be used in the captions. If None, a trigger word will not be used.
64754
- * If no captions are provide the trigger_work will be used instead of captions. If captions are provided, the trigger word will replace the `[trigger]` string in the captions.
64755
- */
64756
- trigger_phrase?: string;
64757
- }
64758
-
64759
- export interface FluxLoraPortraitTrainerOutput extends SharedType_9a6 {}
64760
-
64761
- export interface FluxLoraFillInput {
64762
- /**
64763
- * Acceleration
64764
- * @description Acceleration level for image generation. 'regular' balances speed and quality.
64765
- * @default none
64766
- * @enum {string}
64767
- */
64768
- acceleration?: 'none' | 'regular';
64769
- /**
64770
- * Enable Safety Checker
64771
- * @description If set to true, the safety checker will be enabled.
64772
- * @default true
64773
- */
64774
- enable_safety_checker?: boolean;
64775
- fill_image?: Components.ImageFillInput_1;
64776
- /**
64777
- * Guidance scale (CFG)
64778
- * @description The CFG (Classifier Free Guidance) scale is a measure of how close you want
64779
- * the model to stick to your prompt when looking for a related image to show you.
64780
- * @default 30
64781
- */
64782
- guidance_scale?: number;
64783
- /**
64784
- * Image Size
64785
- * @description The size of the generated image.
64786
- */
64787
- image_size?:
64788
- | Components.ImageSize
64789
- | (
64790
- | 'square_hd'
64791
- | 'square'
64792
- | 'portrait_4_3'
64793
- | 'portrait_16_9'
64794
- | 'landscape_4_3'
64795
- | 'landscape_16_9'
64796
- );
64797
- /**
64798
- * Image Url
64799
- * @description URL of image to use for fill operation
64800
- * @example https://storage.googleapis.com/falserverless/flux-lora/example-images/knight.jpeg
64801
- */
64802
- image_url: string;
64803
- /**
64804
- * Loras
64805
- * @description The LoRAs to use for the image generation. You can use any number of LoRAs
64806
- * and they will be merged together to generate the final image.
64807
- * @default []
64808
- */
64809
- loras?: Components.LoraWeight[];
64810
- /**
64811
- * Mask Url
64812
- * @description The mask to area to Inpaint in.
64813
- * @example https://storage.googleapis.com/falserverless/flux-lora/example-images/mask_knight.jpeg
65012
+ * @example https://storage.googleapis.com/falserverless/flux-lora/example-images/mask_knight.jpeg
64814
65013
  */
64815
65014
  mask_url: string;
64816
65015
  /**
@@ -65052,193 +65251,19 @@ export interface FluxKreaTrainerInput extends SharedType_f11 {}
65052
65251
 
65053
65252
  export interface FluxKreaTrainerOutput extends SharedType_baf1 {}
65054
65253
 
65055
- export interface FluxKreaLoraStreamInput extends SharedType_207 {}
65254
+ export interface FluxKreaLoraStreamInput extends SharedType_c5e {}
65056
65255
 
65057
65256
  export interface FluxKreaLoraStreamOutput extends SharedType_7c6 {}
65058
65257
 
65059
- export interface FluxKreaLoraInpaintingInput {
65060
- /**
65061
- * Enable Safety Checker
65062
- * @description If set to true, the safety checker will be enabled.
65063
- * @default true
65064
- */
65065
- enable_safety_checker?: boolean;
65066
- /**
65067
- * Guidance scale (CFG)
65068
- * @description The CFG (Classifier Free Guidance) scale is a measure of how close you want
65069
- * the model to stick to your prompt when looking for a related image to show you.
65070
- * @default 3.5
65071
- */
65072
- guidance_scale?: number;
65073
- /**
65074
- * Image Size
65075
- * @description The size of the generated image.
65076
- */
65077
- image_size?:
65078
- | Components.ImageSize
65079
- | (
65080
- | 'square_hd'
65081
- | 'square'
65082
- | 'portrait_4_3'
65083
- | 'portrait_16_9'
65084
- | 'landscape_4_3'
65085
- | 'landscape_16_9'
65086
- );
65087
- /**
65088
- * Image Url
65089
- * @description URL of image to use for inpainting. or img2img
65090
- * @example https://storage.googleapis.com/falserverless/example_inputs/dog.png
65091
- */
65092
- image_url: string;
65093
- /**
65094
- * Loras
65095
- * @description The LoRAs to use for the image generation. You can use any number of LoRAs
65096
- * and they will be merged together to generate the final image.
65097
- * @default []
65098
- */
65099
- loras?: Components.LoraWeight[];
65100
- /**
65101
- * Mask Url
65102
- * @description The mask to area to Inpaint in.
65103
- * @example https://storage.googleapis.com/falserverless/example_inputs/dog_mask.png
65104
- */
65105
- mask_url: string;
65106
- /**
65107
- * Num Images
65108
- * @description The number of images to generate. This is always set to 1 for streaming output.
65109
- * @default 1
65110
- */
65111
- num_images?: number;
65112
- /**
65113
- * Num Inference Steps
65114
- * @description The number of inference steps to perform.
65115
- * @default 28
65116
- */
65117
- num_inference_steps?: number;
65118
- /**
65119
- * Output Format
65120
- * @description The format of the generated image.
65121
- * @default jpeg
65122
- * @enum {string}
65123
- */
65124
- output_format?: 'jpeg' | 'png';
65125
- /**
65126
- * Prompt
65127
- * @description The prompt to generate an image from.
65128
- * @example A photo of a lion sitting on a stone bench
65129
- */
65130
- prompt: string;
65131
- /**
65132
- * Seed
65133
- * @description The same seed and the same prompt given to the same version of the model
65134
- * will output the same image every time.
65135
- */
65136
- seed?: number;
65137
- /**
65138
- * Strength
65139
- * @description The strength to use for inpainting/image-to-image. Only used if the image_url is provided. 1.0 is completely remakes the image while 0.0 preserves the original.
65140
- * @default 0.85
65141
- */
65142
- strength?: number;
65143
- /**
65144
- * Sync Mode
65145
- * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
65146
- * @default false
65147
- */
65148
- sync_mode?: boolean;
65149
- }
65258
+ export interface FluxKreaLoraInpaintingInput extends SharedType_90c {}
65150
65259
 
65151
65260
  export interface FluxKreaLoraInpaintingOutput extends SharedType_7c6 {}
65152
65261
 
65153
- export interface FluxKreaLoraImageToImageInput {
65154
- /**
65155
- * Enable Safety Checker
65156
- * @description If set to true, the safety checker will be enabled.
65157
- * @default true
65158
- */
65159
- enable_safety_checker?: boolean;
65160
- /**
65161
- * Guidance scale (CFG)
65162
- * @description The CFG (Classifier Free Guidance) scale is a measure of how close you want
65163
- * the model to stick to your prompt when looking for a related image to show you.
65164
- * @default 3.5
65165
- */
65166
- guidance_scale?: number;
65167
- /**
65168
- * Image Size
65169
- * @description The size of the generated image.
65170
- */
65171
- image_size?:
65172
- | Components.ImageSize
65173
- | (
65174
- | 'square_hd'
65175
- | 'square'
65176
- | 'portrait_4_3'
65177
- | 'portrait_16_9'
65178
- | 'landscape_4_3'
65179
- | 'landscape_16_9'
65180
- );
65181
- /**
65182
- * Image Url
65183
- * @description URL of image to use for inpainting. or img2img
65184
- * @example https://storage.googleapis.com/falserverless/example_inputs/dog.png
65185
- */
65186
- image_url: string;
65187
- /**
65188
- * Loras
65189
- * @description The LoRAs to use for the image generation. You can use any number of LoRAs
65190
- * and they will be merged together to generate the final image.
65191
- * @default []
65192
- */
65193
- loras?: Components.LoraWeight[];
65194
- /**
65195
- * Num Images
65196
- * @description The number of images to generate. This is always set to 1 for streaming output.
65197
- * @default 1
65198
- */
65199
- num_images?: number;
65200
- /**
65201
- * Num Inference Steps
65202
- * @description The number of inference steps to perform.
65203
- * @default 28
65204
- */
65205
- num_inference_steps?: number;
65206
- /**
65207
- * Output Format
65208
- * @description The format of the generated image.
65209
- * @default jpeg
65210
- * @enum {string}
65211
- */
65212
- output_format?: 'jpeg' | 'png';
65213
- /**
65214
- * Prompt
65215
- * @description The prompt to generate an image from.
65216
- * @example A photo of a lion sitting on a stone bench
65217
- */
65218
- prompt: string;
65219
- /**
65220
- * Seed
65221
- * @description The same seed and the same prompt given to the same version of the model
65222
- * will output the same image every time.
65223
- */
65224
- seed?: number;
65225
- /**
65226
- * Strength
65227
- * @description The strength to use for inpainting/image-to-image. Only used if the image_url is provided. 1.0 is completely remakes the image while 0.0 preserves the original.
65228
- * @default 0.85
65229
- */
65230
- strength?: number;
65231
- /**
65232
- * Sync Mode
65233
- * @description If `True`, the media will be returned as a data URI and the output data won't be available in the request history.
65234
- * @default false
65235
- */
65236
- sync_mode?: boolean;
65237
- }
65262
+ export interface FluxKreaLoraImageToImageInput extends SharedType_49b {}
65238
65263
 
65239
65264
  export interface FluxKreaLoraImageToImageOutput extends SharedType_7c6 {}
65240
65265
 
65241
- export interface FluxKreaLoraInput extends SharedType_207 {}
65266
+ export interface FluxKreaLoraInput extends SharedType_c5e {}
65242
65267
 
65243
65268
  export interface FluxKreaLoraOutput extends SharedType_7c6 {}
65244
65269
 
@@ -66977,7 +67002,7 @@ export interface FluxControlLoraDepthImageToImageInput extends SharedType_dce {}
66977
67002
 
66978
67003
  export interface FluxControlLoraDepthImageToImageOutput extends SharedType_7c6 {}
66979
67004
 
66980
- export interface FluxControlLoraDepthInput extends SharedType_fa0 {}
67005
+ export interface FluxControlLoraDepthInput extends SharedType_4c9 {}
66981
67006
 
66982
67007
  export interface FluxControlLoraDepthOutput extends SharedType_7c6 {}
66983
67008
 
@@ -66985,7 +67010,7 @@ export interface FluxControlLoraCannyImageToImageInput extends SharedType_dce {}
66985
67010
 
66986
67011
  export interface FluxControlLoraCannyImageToImageOutput extends SharedType_7c6 {}
66987
67012
 
66988
- export interface FluxControlLoraCannyInput extends SharedType_fa0 {}
67013
+ export interface FluxControlLoraCannyInput extends SharedType_4c9 {}
66989
67014
 
66990
67015
  export interface FluxControlLoraCannyOutput extends SharedType_7c6 {}
66991
67016
 
@@ -77651,8 +77676,8 @@ export interface CodeformerOutput {
77651
77676
  /**
77652
77677
  * @description The generated image file info.
77653
77678
  * @example {
77654
- * "file_size": 423052,
77655
77679
  * "height": 512,
77680
+ * "file_size": 423052,
77656
77681
  * "file_name": "36d3ca4791a647678b2ff01a35c87f5a.png",
77657
77682
  * "content_type": "image/png",
77658
77683
  * "url": "https://storage.googleapis.com/falserverless/model_tests/codeformer/codeformer_restored_1.jpeg",
@@ -79228,6 +79253,11 @@ export interface BytedanceSeedanceV1ProTextToVideoInput {
79228
79253
  * @example true
79229
79254
  */
79230
79255
  enable_safety_checker?: boolean;
79256
+ /**
79257
+ * Num Frames
79258
+ * @description The number of frames to generate. If provided, will override duration.
79259
+ */
79260
+ num_frames?: number;
79231
79261
  /**
79232
79262
  * Prompt
79233
79263
  * @description The text prompt used to generate the video
@@ -79303,6 +79333,11 @@ export interface BytedanceSeedanceV1ProImageToVideoInput {
79303
79333
  * @example https://storage.googleapis.com/falserverless/example_inputs/seedance_pro_i2v_img.jpg
79304
79334
  */
79305
79335
  image_url: string;
79336
+ /**
79337
+ * Num Frames
79338
+ * @description The number of frames to generate. If provided, will override duration.
79339
+ */
79340
+ num_frames?: number;
79306
79341
  /**
79307
79342
  * Prompt
79308
79343
  * @description The text prompt used to generate the video
@@ -79367,6 +79402,11 @@ export interface BytedanceSeedanceV1ProFastTextToVideoInput {
79367
79402
  * @example true
79368
79403
  */
79369
79404
  enable_safety_checker?: boolean;
79405
+ /**
79406
+ * Num Frames
79407
+ * @description The number of frames to generate. If provided, will override duration.
79408
+ */
79409
+ num_frames?: number;
79370
79410
  /**
79371
79411
  * Prompt
79372
79412
  * @description The text prompt used to generate the video
@@ -79437,6 +79477,11 @@ export interface BytedanceSeedanceV1ProFastImageToVideoInput {
79437
79477
  * @example https://storage.googleapis.com/falserverless/example_inputs/seedance_fast_i2v_input.png
79438
79478
  */
79439
79479
  image_url: string;
79480
+ /**
79481
+ * Num Frames
79482
+ * @description The number of frames to generate. If provided, will override duration.
79483
+ */
79484
+ num_frames?: number;
79440
79485
  /**
79441
79486
  * Prompt
79442
79487
  * @description The text prompt used to generate the video
@@ -79501,6 +79546,11 @@ export interface BytedanceSeedanceV1LiteTextToVideoInput {
79501
79546
  * @example true
79502
79547
  */
79503
79548
  enable_safety_checker?: boolean;
79549
+ /**
79550
+ * Num Frames
79551
+ * @description The number of frames to generate. If provided, will override duration.
79552
+ */
79553
+ num_frames?: number;
79504
79554
  /**
79505
79555
  * Prompt
79506
79556
  * @description The text prompt used to generate the video
@@ -79551,6 +79601,11 @@ export interface BytedanceSeedanceV1LiteReferenceToVideoInput {
79551
79601
  * @example true
79552
79602
  */
79553
79603
  enable_safety_checker?: boolean;
79604
+ /**
79605
+ * Num Frames
79606
+ * @description The number of frames to generate. If provided, will override duration.
79607
+ */
79608
+ num_frames?: number;
79554
79609
  /**
79555
79610
  * Prompt
79556
79611
  * @description The text prompt used to generate the video
@@ -79635,6 +79690,11 @@ export interface BytedanceSeedanceV1LiteImageToVideoInput {
79635
79690
  * @example https://fal.media/files/koala/f_xmiodPjhiKjdBkFmTu1.png
79636
79691
  */
79637
79692
  image_url: string;
79693
+ /**
79694
+ * Num Frames
79695
+ * @description The number of frames to generate. If provided, will override duration.
79696
+ */
79697
+ num_frames?: number;
79638
79698
  /**
79639
79699
  * Prompt
79640
79700
  * @description The text prompt used to generate the video
@@ -81071,8 +81131,8 @@ export interface BagelEditOutput {
81071
81131
  * @description The edited images.
81072
81132
  * @example [
81073
81133
  * {
81074
- * "file_size": 423052,
81075
81134
  * "height": 1024,
81135
+ * "file_size": 423052,
81076
81136
  * "file_name": "hQnndOMvGSt2UsYAiV3vs.jpeg",
81077
81137
  * "content_type": "image/jpeg",
81078
81138
  * "url": "https://storage.googleapis.com/falserverless/bagel/hQnndOMvGSt2UsYAiV3vs.jpeg",
@@ -81134,8 +81194,8 @@ export interface BagelOutput {
81134
81194
  * @description The generated images.
81135
81195
  * @example [
81136
81196
  * {
81137
- * "file_size": 423052,
81138
81197
  * "height": 1024,
81198
+ * "file_size": 423052,
81139
81199
  * "file_name": "wRhCPSyiKTiLnnWvUpGIl.jpeg",
81140
81200
  * "content_type": "image/jpeg",
81141
81201
  * "url": "https://storage.googleapis.com/falserverless/bagel/wRhCPSyiKTiLnnWvUpGIl.jpeg",
@@ -81205,7 +81265,7 @@ export interface AutoCaptionInput {
81205
81265
  txt_font?: string;
81206
81266
  /**
81207
81267
  * Video Url
81208
- * @description URL to the .mp4 video with audio. Only videos of size <100MB are allowed.
81268
+ * @description URL to the .mp4 video with audio. Only videos of size <400MB are allowed.
81209
81269
  */
81210
81270
  video_url: string;
81211
81271
  }
@@ -82774,8 +82834,8 @@ export interface ClarityaiCrystalVideoUpscalerOutput {
82774
82834
  * "height": 2160,
82775
82835
  * "duration": 13.056527,
82776
82836
  * "url": "https://storage.googleapis.com/falserverless/example_outputs/crystal_upscaler/video_upscaling/video_out.mp4",
82777
- * "width": 4096,
82778
82837
  * "fps": 23.130193905817176,
82838
+ * "width": 4096,
82779
82839
  * "file_name": "w0VQQvPdwvV2GSCtRTMzh_hDH8SPrB.mp4",
82780
82840
  * "num_frames": 302,
82781
82841
  * "content_type": "video/mp4"
@@ -83744,6 +83804,7 @@ export interface BriaFiboEditEditInput {
83744
83804
  * @default
83745
83805
  */
83746
83806
  negative_prompt?: string;
83807
+ original_vgl?: Components.StructuredInstruction;
83747
83808
  /**
83748
83809
  * Seed
83749
83810
  * @description Random seed for reproducibility.
@@ -83931,19 +83992,19 @@ export interface BriaEmbedProductInput {
83931
83992
  * @example [
83932
83993
  * {
83933
83994
  * "coordinates": {
83934
- * "height": 300,
83935
83995
  * "y": 317,
83936
- * "width": 100,
83937
- * "x": 300
83996
+ * "height": 300,
83997
+ * "x": 300,
83998
+ * "width": 100
83938
83999
  * },
83939
84000
  * "image_source": "https://bria-datasets.s3.us-east-1.amazonaws.com/embed-product/a_standing_lamp_over_white_background_0.png"
83940
84001
  * },
83941
84002
  * {
83942
84003
  * "coordinates": {
83943
- * "height": 156,
83944
84004
  * "y": 287,
83945
- * "width": 120,
83946
- * "x": 646
84005
+ * "height": 156,
84006
+ * "x": 646,
84007
+ * "width": 120
83947
84008
  * },
83948
84009
  * "image_source": "https://bria-datasets.s3.us-east-1.amazonaws.com/embed-product/a_wall_picture_on_white_background_0.png"
83949
84010
  * }