GET
/v1/schemas/fal/video/fal-ai/ltxv-13b-098-distilled/multiconditioning
Accept:
| field | recorded source |
|---|---|
| schema | Source linked above. upstream-spec Source trace6b54d3fafe18726d0185aaa3fdb4e127fa11043a6baf66e4b578bbfced79ff6f{
"fetchedAt": 1793250001,
"extractorVersion": "1",
"verifiedAt": null
} |
Request body
20 properties · 1 required · 3 $defs| property | type | description | constraints |
|---|---|---|---|
| prompt* | string | Text prompt to guide generation | |
| videos | VideoConditioningInput[] | Videos to use as conditioning | default: [] |
| tone_map_compression_ratio | number | 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. | 0 ≤ n ≤ 1 · default: 0 |
| seed | integer | Random seed for generation | nullable |
| resolution | enum | Resolution of the generated video. | 480p | 720p · default: "720p" |
| enable_detail_pass | boolean | 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. | default: false |
| num_frames | integer | The number of frames in the video. | 9 ≤ n ≤ 1441 · default: 121 |
| second_pass_num_inference_steps | integer | Number of inference steps during the second pass. | 2 ≤ n ≤ 12 · default: 8 |
| reverse_video | boolean | Whether to reverse the video. | default: false |
| second_pass_skip_initial_steps | integer | 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. | 1 ≤ n ≤ 11 · default: 5 |
| images | ImageConditioningInput[] | URL of images to use as conditioning | default: [] |
| aspect_ratio | enum | The aspect ratio of the video. | 9:16 | 1:1 | 16:9 | auto · default: "auto" |
| frame_rate | integer | The frame rate of the video. | 1 ≤ n ≤ 60 · default: 24 |
| negative_prompt | string | Negative prompt for generation | default: "worst quality, inconsistent mot… |
| constant_rate_factor | integer | 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. | 0 ≤ n ≤ 51 · default: 29 |
| enable_prompt_expansion | boolean | Whether to expand the prompt using a language model. | default: false |
| enable_safety_checker | boolean | Whether to enable the safety checker. Disabling it requires account authorization; unauthorized requests are always checked. | default: true |
| loras | LoRAWeight[] | LoRA weights to use for generation | default: [] |
| temporal_adain_factor | number | 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 va… | 0 ≤ n ≤ 1 · default: 0.5 |
| first_pass_num_inference_steps | integer | Number of inference steps during the first pass. | 2 ≤ n ≤ 12 · default: 8 |
Types
3 $defsImageConditioningInputobject
| image_url* | string | URL of image to use as conditioning | |
| start_frame_num | integer | Frame number of the image from which the conditioning starts. Must be a multiple of 8. | 0 ≤ n ≤ 1440 · default: 0 |
| strength | number | Strength of the conditioning. 0.0 means no conditioning, 1.0 means full conditioning. | 0 ≤ n ≤ 1 · default: 1 |
LoRAWeightobject
| path* | string | URL or path to the LoRA weights. | |
| weight_name | string | Name of the LoRA weight. Only used if `path` is a HuggingFace repository, and is only required when the repository contains multiple LoRA weights. | nullable |
| scale | number | Scale of the LoRA weight. This is a multiplier applied to the LoRA weight when loading it. | 0 ≤ n ≤ 4 · default: 1 |
VideoConditioningInputobject
| max_num_frames | integer | Maximum number of frames to use from the video. If None, all frames will be used. | 1 ≤ n ≤ 1441 · default: 1441 |
| video_url* | string | URL of video to use as conditioning | |
| resample_fps | boolean | Whether to resample the video to a specific FPS. If True, the `target_fps` parameter will be used to resample the video. | default: false |
| preprocess | boolean | Whether to preprocess the video. If True, the video will be preprocessed to match the conditioning type. This is a no-op for RGB conditioning. | default: false |
| limit_num_frames | boolean | Whether to limit the number of frames used from the video. If True, the `max_num_frames` parameter will be used to limit the number of frames. | default: false |
| target_fps | integer | Target FPS to resample the video to. Only relevant if `resample_fps` is True. | 1 ≤ n ≤ 60 · default: 24 |
| start_frame_num | integer | Frame number of the video from which the conditioning starts. Must be a multiple of 8. | 0 ≤ n ≤ 1440 · default: 0 |
| reverse_video | boolean | Whether to reverse the video. This is useful for tasks where the video conditioning should be applied in reverse order. | default: false |
| strength | number | Strength of the conditioning. 0.0 means no conditioning, 1.0 means full conditioning. | 0 ≤ n ≤ 1 · default: 1 |
| conditioning_type | enum | Type of conditioning this video provides. This is relevant to ensure in-context LoRA weights are applied correctly, as well as selecting the correct preprocessing pipeline, when enabled. | rgb | depth | pose | canny · nullable |
Validate a payload
before you spend tokensshell
$ curl -X POST https://modelschemas.com/v1/validate \
-d '{"provider":"fal","endpointId":"fal-ai/ltxv-13b-098-distilled/multiconditioning","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}