This API transforms textual descriptions into dynamic videos. By interpreting and visualizing the input text, it creates engaging video content that corresponds to the described scenarios, events, or narratives. This capability is ideal for content creation where the visual representation of text-based information enhances understanding or entertainment value.
This is an asynchronous API; only the task_id will be returned. You should use the task_id to request the Task Result API to retrieve the video generation results.
Control the data content of the mock event. When set to TASK_STATUS_SUCCEED, you’ll receive a normal response; when set to TASK_STATUS_FAILED, you’ll receive an error response.
Enum: TASK_STATUS_SUCCEED, TASK_STATUS_FAILED
Name of SD1.x checkpoints; this parameter specifies the name of the model checkpoint. Retrieve the corresponding sd_name value by invoking the Query Model API with filter.types=checkpoint as the query parameter.
This setting determines how closely Stable Diffusion will adhere to your prompt. Higher guidance forces the model to better follow the prompt but may result in lower quality output. Range [1, 30].
A seed is a number from which Stable Diffusion generates noise, which, makes generation deterministic. Using the same seed and set of parameters will produce identical content each time, minimum -1. Defaults to -1.
LoRA is a fast and lightweight training method that inserts and trains a significantly smaller number of parameters instead of all the model parameters. Currently supports up to 5 LoRAs.
Textual Inversion is a training method for personalizing models by learning new text embeddings from a few example images, currently supports up to 5 embeddings.
Name of textual Inversion model, you can call the Get Model API endpoint with parameter filter.types=textualinversion to retrieve the sd_name_in_api field as the model_name.
The closed_loop parameter controls the behavior of an animation when it loops. Specifically, it determines whether the last frame of the animation will transition smoothly back to the first frame.
This parameter indicates the number of layers to stop from the bottom during optimization, so clip_skip on 2 would mean, that in SD1.x model where the CLIP has 12 layers, you would stop at 10th layer, Range [1, 12], get reference at A brief introduction to Clip Skip.
This API helps generate videos from text. The returned video can be accessed via the API /v3/async/task-result using the task_id.Try it in playground.Request:
Copy
curl --location 'https://api.novita.ai/v3/async/txt2video' \--header 'Authorization: Bearer {{API Key}}' \--header 'Content-Type: application/json' \--data '{ "model_name": "darkSushiMixMix_225D_64380.safetensors", "height": 512, "width": 512, "steps": 20, "seed": -1, "prompts": [ { "frames": 32, "prompt": "In the wintry dusk, a little girl holds matches tightly." }, { "frames": 32, "prompt": "A little girl, barefoot on the frosty pavement, seeks solace." }, { "frames": 32, "prompt": "A little girl with each match experiences a fleeting dance of warmth and hope." }, { "frames": 32, "prompt": "In the quiet night, a little girl's silent story unfolds." } ], "negative_prompt": "nsfw, ng_deepnegative_v1_75t, badhandv4, (worst quality:2), (low quality:2), (normal quality:2), lowres, ((monochrome)), ((grayscale)), watermark"}'
Use task_id to get imagesHTTP status codes in the 2xx range indicate that the request has been successfully accepted, while status codes in the 5xx range indicate internal server errors.You can get videos url in videos of response.Request:
Copy
curl --location --request GET 'https://api.novita.ai/v3/async/task-result?task_id=fa20dff3-18cb-4417-a7f8-269456a35154' \--header 'Authorization: Bearer {{API Key}}'