Bach Video Generation

Bach Video Generation

Bach video generation API generates high-quality videos from text, keyframes, or reference assets. It is suited for content creation, film previews, brand promotion, e-commerce advertising, and other scenarios requiring flexible control over visuals and style.

Three generation modes are currently supported (specified by input.task):

  • Text to Video (t2v): Generate a video directly from a text description; this is the default when task is omitted
  • First/Last Frame (i2v): Specify the opening and/or closing frame to generate a smoothly transitioned video
  • Multimodal Reference (r2v): Combine image, video, and audio references to preserve visual subjects, camera rhythm, and sound atmosphere

Reference assets must be publicly accessible HTTP/HTTPS URLs. Base64 is not supported.

  • Base URL: https://api-gen-na.bach.art/api/vdr
  • Authentication: API Key (Bearer Token). Include Authorization: Bearer <API_KEY> in the request header

Create Generation Task

Endpoint Information

  • Protocol: HTTPS
  • Request Path: /video/v2
  • Request Method: POST
  • Request Format: application/json
  • Response Format: application/json

Request Headers

Content-Type

Type: string ｜ Required: Yes ｜ Default: None

Data exchange format.

  • Fixed value: application/json

Authorization

Type: string ｜ Required: Yes ｜ Default: None

Authentication information, Bearer Token scheme.

  • Format: Bearer <API_KEY>
  • [Note] There is a single space between Bearer and <API_KEY>.

Request Body

model_name

Type: string ｜ Required: Yes

Set the model name which is used to generate the video.

enhance_prompt

Type: bool ｜ Required: No ｜ Default: true

Whether to enhance the prompt. When true: the prompt is enhanced When false: the prompt is not enhanced

callback_url

Type: string ｜ Required: No ｜ Default: None

Callback URL notified with the result of this task

  • Must be a valid HTTP or HTTPS URL.
  • Must not exceed 500 characters.
  • If configured, the server proactively sends notifications whenever the task status changes.

input

Type: object ｜ Required: Yes ｜ Default: None

Video generation parameters.

input Properties

task

Type: string ｜ Required: No ｜ Default: t2v

Generation mode. Enum values: t2v, i2v, r2v.

prompt

Type: string ｜ Required: Yes ｜ Default: None

Positive text prompt, i.e., the description of the video content.

  • Must not exceed 6,500 characters.

duration

Type: integer ｜ Required: No ｜ Default: 8

Duration of the generated video, in seconds (s). Value range: 5 ~ 15.

generate_audio

Type: bool ｜ Required: No ｜ Default: false

Whether to generate audio along with the video.

  • When true: synchronized audio generation is enabled
  • When false: synchronized audio generation is disabled

aspect_ratio

Type: string ｜ Required: No ｜ Default: See description

Aspect ratio of the generated video (width:height). Enum values: 16:9, 9:16, 1:1, auto.

  • Text to Video (t2v), Multimodal Reference (r2v): defaults to 16:9 when omitted.
  • First/Last Frame (i2v): fixed to auto.

resolution

Type: string ｜ Required: No ｜ Default: 720p

Output resolution. Enum values: 720p, 1080p.

seed

Type: integer ｜ Required: No ｜ Default: None

Random seed. Value range: 1 ~ 100000.

image_url

Type: string ｜ Required: Conditionally required ｜ Default: None

First-frame image URL, the starting reference frame of the generated video.

  • Used only in i2v.
  • Must be a publicly accessible HTTP/HTTPS URL.
  • [Note] If end_image_url is not provided, image_url is required.

end_image_url

Type: string ｜ Required: Conditionally required ｜ Default: None

Last-frame image URL, the closing target frame of the generated video.

  • Used only in i2v.
  • Must be a publicly accessible HTTP/HTTPS URL.
  • [Note] If image_url is not provided, end_image_url is required; when both are provided, the video generates a natural transition between the first and last frames.

reference_inputs

Type: array ｜ Required: Conditionally required ｜ Default: None

Reference assets.

[Important] Prompt references (r2v) Use @Image1 / @Video1 / @Audio1 to reference the items in reference_inputs in the order the assets appear.

<a id="reference_inputs-items"></a>

reference_inputs Items

type

Type: string ｜ Required: Yes ｜ Default: None

Asset type. Enum values: image / video / audio.

url

Type: string ｜ Required: Yes ｜ Default: None

Asset URL.

  • Must be a publicly accessible HTTP/HTTPS URL.

[Important] Multimodal Reference (r2v)

  • reference_inputs must not be empty
  • Optional start_time (in seconds, ≥ 0)

Request Examples

Text to Video (t2v)

bash
Copy
curl --location --request POST 'https://api-gen-na.bach.art/api/vdr/video/v2' \
--header 'Authorization: Bearer YOUR_API_KEY' \
--header 'Content-Type: application/json' \
--data '{
    "model_name": "$PREFERRED_MODEL_NAME",
    "enhance_prompt": true,
    "input": {
      "task": "t2v",
      "prompt": "A cinematic daytime street scene.",
      "duration": 8,
      "generate_audio": false,
      "seed": 1101,
      "aspect_ratio": "16:9",
      "resolution": "720p"
    }
}'

First/Last Frame (i2v)

bash
Copy
curl --location --request POST 'https://api-gen-na.bach.art/api/vdr/video/v2' \
--header 'Authorization: Bearer YOUR_API_KEY' \
--header 'Content-Type: application/json' \
--data '{
    "model_name": "$PREFERRED_MODEL_NAME",
    "input": {
      "task": "i2v",
      "prompt": "Continue naturally between the endpoint frames.",
      "duration": 8,
      "generate_audio": false,
      "seed": 1101,
      "image_url": "https://example.com/first.png",
      "end_image_url": ""
    }
}'

Multimodal Reference (r2v)

bash
Copy
curl --location --request POST 'https://api-gen-na.bach.art/api/vdr/video/v2' \
--header 'Authorization: Bearer YOUR_API_KEY' \
--header 'Content-Type: application/json' \
--data '{
    "model_name": "$PREFERRED_MODEL_NAME",
    "input": {
      "task": "r2v",
      "prompt": "Use @Video1 as motion reference and @Audio1 as audio reference.",
      "duration": 8,
      "seed": 1101,
      "reference_inputs": [
        { "type": "video", "url": "https://example.com/input.mp4" },
        { "type": "audio", "url": "https://example.com/reference.mp3" }
      ]
    }
}'

Response Body

json
Copy
{
  "code": 200,                                            // Status code; 200 on success, non-200 on failure — see message for details
  "data": {
    "task_id": "550e8400-e29b-41d4-a716-446655440000",    // Task ID, used to query the result
    "status": "TASK_PENDING",                             // Task status; for enum values see "Task Status Enum Values"
    "created_at": 1778313600                              // Task creation time, Unix timestamp, in seconds
  },
  "timestamp": 1778313600                                 // Response time, Unix timestamp, in seconds
}

code

Type: integer

Status code. 200 on success; non-200 on failure, with the specific reason in message.

message

Type: string

Result description. Typically omitted on success (code: 200).

task_id

Type: string

Task ID, used to query the result.

status

Type: string

Task status; for enum values see Task Status Enum Values.

created_at

Type: number

Task creation time, Unix timestamp, in seconds.

timestamp

Type: number

Response time, Unix timestamp, in seconds.

Query Task (Single)

Endpoint Information

  • Protocol: HTTPS
  • Request Path: /video/v2/{task_id}
  • Request Method: GET
  • Request Format: application/json
  • Response Format: application/json

Queries the processing progress and result of the specified task. On successful generation, the video download URL is returned.

Request Headers

Content-Type

Type: string ｜ Required: Yes ｜ Default: None

Data exchange format.

  • Fixed value: application/json

Authorization

Type: string ｜ Required: Yes ｜ Default: None

Authentication information, Bearer Token scheme.

  • Format: Bearer <API_KEY>
  • [Note] There is a single space between Bearer and <API_KEY>.

Path Parameters

task_id

Type: string ｜ Required: Yes ｜ Default: None

The task ID returned when the task was created.

  • [Note] This is a path parameter; place the value directly in the request path.

Request Example

bash
Copy
curl --location --request GET 'https://api-gen-na.bach.art/api/vdr/video/v2/550e8400-e29b-41d4-a716-446655440000' \
--header 'Authorization: Bearer YOUR_API_KEY'

Response Body

json
Copy
{
  "code": 200,                                            // Status code; 200 on success
  "data": {
    "task_id": "550e8400-e29b-41d4-a716-446655440000",    // Task ID
    "status": "TASK_SUCCEEDED",                           // Task status; for enum values see "Task Status Enum Values"
    "created_at": 1778313600,                             // Task creation time, Unix timestamp, in seconds
    "started_at": 1778313610,                             // Processing start time, Unix timestamp, in seconds
    "completed_at": 1778313700,                           // Completion time, Unix timestamp, in seconds
    "video_url": "https://example.com/result.mp4",        // Download URL of the generated video; returned only when status is TASK_SUCCEEDED
    "usage": {                                            // Usage; returned only in terminal states
      "video_output": 5                                   // Duration of this video in seconds; equals the duration at creation on success, 0 on failure
    }
  },
  "timestamp": 1778313700                                 // Response time, Unix timestamp, in seconds
}

task_id

Type: string

Task ID.

status

Type: string

Task status; for enum values see Task Status Enum Values.

created_at

Type: number

Task creation time, Unix timestamp, in seconds.

started_at

Type: number

Processing start time, Unix timestamp, in seconds.

completed_at

Type: number

Completion time, Unix timestamp, in seconds.

video_url

Type: string

Download URL of the generated video.

  • [Note] Returned only when the status is TASK_SUCCEEDED.

usage

Type: object

Usage.

  • Returned only in terminal states.
  • video_output is the duration of this video in seconds; it equals the duration specified at creation on success, and 0 on failure.

Task Status Enum Values

  • TASK_PENDING: Task submitted, awaiting processing
  • TASK_PROCESSING: Task is being processed
  • TASK_SUCCEEDED: Video generated successfully
  • TASK_FAILED: Video generation failed

Appendix: Authentication and Token Generation

All API requests must include Authorization in the request header.

text
Copy
Authorization: Bearer <API_TOKEN>

API_TOKEN is generated from an AccessKey and a SecretKey. The JWT uses the HS256 signing algorithm and follows the RFC 7519 standard.

Generating a Token

The following example is written in Python and demonstrates how to generate an API Token using the industry-standard JWT Bearer Token approach.

python
Copy
# Python example: generate a JWT Bearer Token
import jwt
from datetime import datetime, timedelta

ak = ""  # Enter your AccessKey
sk = ""  # Enter your SecretKey

def generate_token(ak: str, sk: str) -> str:
    now = datetime.now()

    headers = {
        "alg": "HS256",
        "typ": "JWT",
    }

    payload = {
        "iss": ak,
        "nbf": int((now - timedelta(seconds=5)).timestamp()),
        "exp": int((now + timedelta(days=10)).timestamp()),
    }

    return jwt.encode(payload, sk.encode("utf-8"), headers=headers)

api_token = generate_token(ak, sk)
print(api_token)

Security Recommendations

  • Never expose your SecretKey in browsers, mobile clients, or public repositories.
  • Generate the Token on your server side, and have the server make requests to the Bach API.
  • If authentication fails, first check whether the Token has expired, whether the signature is correct, and whether there is a single space between Bearer and the Token.
  • Version: v2.0
  • Last Updated: 2026-09-04
  • © 2026 Video Rebirth team. All rights reserved.
Previous
Reference to Video
Next
Text to Image
On this page
Bach Video Generation | bach.art | bach.art