> ## Documentation Index
> Fetch the complete documentation index at: https://apidoc.cometapi.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Tạo video Flux 3
> Tạo video Flux 3 từ văn bản, hình ảnh tham chiếu hoặc khung hình đầu và cuối với preset kích thước 720p hoặc 1080p.
Tạo video Flux 3 từ văn bản, hình ảnh tham chiếu hoặc khung hình đầu và cuối. Lưu `id` được trả về để thăm dò tác vụ và tải video đã hoàn tất xuống.
`POST /v1/videos` sử dụng `multipart/form-data`. Gửi các tùy chọn điều khiển dưới dạng trường biểu mẫu. Đối với dữ liệu đầu vào hình ảnh, hãy dùng URL hình ảnh tham chiếu hoặc khung hình mở đầu và kết thúc.
Giữ `input_reference` tách biệt với `first_frame` và `last_frame`.
## Chọn chế độ đầu vào
| Mục tiêu | Trường bắt buộc | Trường tùy chọn |
| --------------------------------- | ----------------------------------------------------------- | ----------------- |
| Video từ văn bản | `model=flux-3`, `prompt` | `seconds`, `size` |
| Video từ hình ảnh HTTPS | `model=flux-3`, `prompt`, một URL `input_reference` | `seconds`, `size` |
| Các keyframe theo thứ tự | `model=flux-3`, `prompt`, các URL `input_reference` lặp lại | `seconds`, `size` |
| Khung hình mở đầu đã tải lên | `model=flux-3`, `prompt`, một tệp `first_frame` | `seconds`, `size` |
| Video dùng khung hình đầu và cuối | `model=flux-3`, `prompt`, `first_frame`, `last_frame` | `seconds`, `size` |
## Sử dụng hình ảnh tham chiếu
Trường `input_reference` chỉ chấp nhận các URL hình ảnh HTTPS có thể truy cập công khai. Gửi trường này một lần cho một hình ảnh hoặc lặp lại theo thứ tự keyframe cho nhiều hình ảnh. Dùng một URL cho mỗi trường, tối đa 10 URL.
Mô tả chuyển động, hành vi của camera và các chi tiết trực quan mà video được tạo cần giữ lại từ hình ảnh tham chiếu.
Các mẫu yêu cầu ở đầu trang này minh họa dữ liệu đầu vào một hình ảnh và nhiều hình ảnh. Yêu cầu sau đây sử dụng hai hình ảnh theo thứ tự đã gửi:
```bash theme={null}
curl \
https://api.cometapi.com/v1/videos \
-H "Authorization: Bearer $COMETAPI_KEY" \
--form-string 'model=flux-3' \
--form-string 'prompt=Move smoothly through the supplied reference images in their given order. Preserve recognizable subjects and visual details as the scene changes. Keep the camera movement gentle.' \
--form-string 'seconds=5' \
--form-string 'size=1280x720' \
--form-string 'input_reference=https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png' \
--form-string 'input_reference=https://apidoc.cometapi.com/images/image/gemini/6100640_569420.png'
```
## Đặt khung hình đầu và cuối
Gửi `first_frame` để đặt hình ảnh mở đầu. Để đồng thời đặt hình ảnh kết thúc, hãy thêm `last_frame` và mô tả chuyển động giữa hai khung hình trong `prompt`.
Đối với chỉ khung hình mở đầu, hãy bỏ qua `last_frame`. Khung hình kết thúc yêu cầu cả `first_frame` và `last_frame`.
Mỗi trường chấp nhận một URL HTTPS có thể truy cập công khai hoặc một tệp PNG hoặc JPEG. Gửi mỗi trường một lần, bằng URL hoặc tệp. Giữ mỗi tệp ở mức không quá 20 MiB (20 × 1024 × 1024 byte).
Các ví dụ về khung hình ở đầu trang này minh họa dữ liệu đầu vào bằng URL và tải tệp lên, với `seconds` và `size` được đặt tường minh.
## Đặt thời lượng và kích thước
Đặt `seconds` một cách tường minh thành số nguyên từ `5` đến `20`.
Đặt `size` thành một trong các giá trị preset `WxH` chính xác sau đây:
| Cấp độ độ phân giải | Kích thước |
| ------------------- | ----------- |
| `720p` | `1280x720` |
| `1080p` | `1920x1080` |
Giá trị kích thước chọn cấp độ độ phân giải đầu ra. Đối với video từ văn bản, kích thước được mã hóa có thể được điều chỉnh để căn chỉnh codec. Đối với video từ hình ảnh, hình ảnh tham chiếu cũng có thể xác định khung hình và tỷ lệ khung hình cuối cùng.
## Luồng tác vụ
Gửi yêu cầu biểu mẫu multipart và lưu `id` được trả về.
Gọi [Truy xuất video Flux 3](./retrieve) cho đến khi `status` là `completed` hoặc `failed`.
Khi tác vụ ở trạng thái `completed`, hãy gọi [Tải nội dung video Flux 3 xuống](./retrieve-content) để lưu tệp MP4.
## OpenAPI
````yaml api/openapi/video/flux-3/post-create.openapi.json POST /v1/videos
openapi: 3.1.0
info:
title: Flux 3 Video Create API
version: 1.0.0
description: >-
Create a Flux 3 video from text, a reference image, or opening and closing
frames with multipart form data.
servers:
- url: https://api.cometapi.com
security:
- bearerAuth: []
paths:
/v1/videos:
post:
summary: Create a Flux 3 video task
description: >-
Create a Flux 3 video from text, one reference image URL, or opening and
closing frames.
operationId: flux_3_create_video
requestBody:
required: true
content:
multipart/form-data:
schema:
$ref: '#/components/schemas/Flux3CreateRequest'
encoding:
input_reference:
contentType: text/plain
style: form
explode: true
first_frame:
contentType: image/png, image/jpeg, text/plain
last_frame:
contentType: image/png, image/jpeg, text/plain
examples:
text_to_video:
summary: Text-to-video
value:
model: flux-3
prompt: >-
A paper boat glides across a still pond while the camera
moves forward.
seconds: 5
size: 1280x720
https_image_to_video:
summary: Image-to-video with an HTTPS image
value:
model: flux-3
prompt: >-
Animate the reference scene with a slow camera move and
natural motion.
seconds: 5
size: 1280x720
input_reference: >-
https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png
first_last_frame_urls:
summary: First and last frames from URLs
value:
model: flux-3
prompt: Move smoothly from the opening to the closing scene.
seconds: 5
size: 1280x720
first_frame: https://your-image-host/first-frame.png
last_frame: https://your-image-host/last-frame.png
first_last_frame_files:
summary: First and last frames from files
value:
model: flux-3
prompt: Move smoothly from the opening to the closing scene.
seconds: 5
size: 1280x720
first_frame: '@/path/to/first-frame.png'
last_frame: '@/path/to/last-frame.png'
opening_frame_file:
summary: Opening frame from a file
value:
model: flux-3
prompt: >-
Animate the opening frame with a slow camera move and
natural motion.
seconds: 5
size: 1280x720
first_frame: '@/path/to/reference.png'
input_reference_url_pair:
summary: Ordered keyframes from image URLs
value:
model: flux-3
prompt: >-
Move smoothly through the supplied reference images in their
given order. Preserve recognizable subjects and visual
details as the scene changes. Keep the camera movement
gentle.
seconds: 5
size: 1280x720
input_reference:
- >-
https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png
- >-
https://apidoc.cometapi.com/images/image/gemini/6100640_569420.png
responses:
'200':
description: >-
Task created. Store the returned id and poll GET
/v1/videos/{task_id}.
content:
application/json:
schema:
$ref: '#/components/schemas/Flux3VideoTask'
example:
id:
task_id:
object: video
model: flux-3
status: queued
progress: 0
created_at: 1779938152
'400':
description: >-
The request is missing a required field or contains an unsupported
value.
'401':
description: The API key is missing or invalid.
security:
- bearerAuth: []
x-codeSamples:
- lang: Shell
label: Text-to-video
source: |-
curl \
https://api.cometapi.com/v1/videos \
-H "Authorization: Bearer $COMETAPI_KEY" \
--form-string 'model=flux-3' \
--form-string 'prompt=A paper boat glides across a still pond while the camera moves forward.' \
--form-string 'seconds=5' \
--form-string 'size=1280x720'
- lang: Python
label: Text-to-video
source: |
import os
import requests
fields = [
("model", (None, "flux-3")),
(
"prompt",
(
None,
"A paper boat glides across a still pond while the "
"camera moves forward.",
),
),
("seconds", (None, "5")),
("size", (None, "1280x720")),
]
response = requests.post(
"https://api.cometapi.com/v1/videos",
headers={
"Authorization": "Bearer " + os.environ["COMETAPI_KEY"]
},
files=fields,
timeout=120,
)
response.raise_for_status()
print(response.json())
- lang: JavaScript
label: Text-to-video
source: |
const form = new FormData();
form.append("model", "flux-3");
form.append(
"prompt",
"A paper boat glides across a still pond while the camera moves forward.",
);
form.append("seconds", "5");
form.append("size", "1280x720");
const response = await fetch("https://api.cometapi.com/v1/videos", {
method: "POST",
headers: { Authorization: `Bearer ${process.env.COMETAPI_KEY}` },
body: form,
});
if (!response.ok) {
throw new Error(await response.text());
}
console.log(await response.json());
- lang: Shell
label: HTTPS image-to-video
source: |-
curl \
https://api.cometapi.com/v1/videos \
-H "Authorization: Bearer $COMETAPI_KEY" \
--form-string 'model=flux-3' \
--form-string 'prompt=Animate the reference scene with a slow camera move and natural motion.' \
--form-string 'seconds=5' \
--form-string 'size=1280x720' \
--form-string 'input_reference=https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png'
- lang: Python
label: HTTPS image-to-video
source: |
import os
import requests
fields = [
("model", (None, "flux-3")),
(
"prompt",
(
None,
"Animate the reference scene with a slow camera move and "
"natural motion.",
),
),
("seconds", (None, "5")),
("size", (None, "1280x720")),
(
"input_reference",
(None, "https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png"),
),
]
response = requests.post(
"https://api.cometapi.com/v1/videos",
headers={
"Authorization": "Bearer " + os.environ["COMETAPI_KEY"]
},
files=fields,
timeout=120,
)
response.raise_for_status()
print(response.json())
- lang: JavaScript
label: HTTPS image-to-video
source: >
const form = new FormData();
form.append("model", "flux-3");
form.append(
"prompt",
"Animate the reference scene with a slow camera move and natural motion.",
);
form.append("seconds", "5");
form.append("size", "1280x720");
form.append("input_reference",
"https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png");
const response = await fetch("https://api.cometapi.com/v1/videos", {
method: "POST",
headers: { Authorization: `Bearer ${process.env.COMETAPI_KEY}` },
body: form,
});
if (!response.ok) {
throw new Error(await response.text());
}
console.log(await response.json());
- lang: Shell
label: First and last frames from URLs
source: |-
image1=https://your-image-host/first-frame.png
image2=https://your-image-host/last-frame.png
curl \
https://api.cometapi.com/v1/videos \
-H "Authorization: Bearer $COMETAPI_KEY" \
--form-string model=flux-3 \
--form-string 'prompt=Move smoothly from the opening to the closing scene.' \
--form-string seconds=5 \
--form-string size=1280x720 \
--form-string "first_frame=$image1" \
--form-string "last_frame=$image2"
- lang: Python
label: First and last frames from URLs
source: |
import os
import requests
url1 = "https://your-image-host/first-frame.png"
url2 = "https://your-image-host/last-frame.png"
prompt = "Move smoothly from the opening to the closing scene."
fields = [
("model", (None, "flux-3")),
("prompt", (None, prompt)),
("seconds", (None, "5")),
("size", (None, "1280x720")),
(
"first_frame",
(None, url1),
),
(
"last_frame",
(None, url2),
),
]
response = requests.post(
"https://api.cometapi.com/v1/videos",
headers={"Authorization": "Bearer " + os.environ["COMETAPI_KEY"]},
files=fields,
timeout=120,
)
response.raise_for_status()
print(response.json())
- lang: JavaScript
label: First and last frames from URLs
source: >
const form = new FormData();
form.append("model", "flux-3");
form.append("prompt", "Move smoothly from the opening to the closing
scene.");
form.append("seconds", "5");
form.append("size", "1280x720");
form.append(
"first_frame",
"https://your-image-host/first-frame.png",
);
form.append(
"last_frame",
"https://your-image-host/last-frame.png",
);
const response = await fetch("https://api.cometapi.com/v1/videos", {
method: "POST",
headers: { Authorization: "Bearer " + process.env.COMETAPI_KEY },
body: form,
});
if (!response.ok) {
throw new Error(await response.text());
}
console.log(await response.json());
- lang: Shell
label: First and last frames from files
source: |-
curl \
https://api.cometapi.com/v1/videos \
-H "Authorization: Bearer $COMETAPI_KEY" \
--form-string model=flux-3 \
--form-string 'prompt=Move smoothly from the opening to the closing scene.' \
--form-string seconds=5 \
--form-string size=1280x720 \
--form 'first_frame=@/path/to/first-frame.png;type=image/png' \
--form 'last_frame=@/path/to/last-frame.png;type=image/png'
- lang: Python
label: First and last frames from files
source: |
import os
from contextlib import ExitStack
import requests
prompt = (
"Move smoothly from the opening to the closing scene."
)
with ExitStack() as stack:
image_1 = stack.enter_context(open("/path/to/first-frame.png", "rb"))
image_2 = stack.enter_context(open("/path/to/last-frame.png", "rb"))
fields = [
("model", (None, "flux-3")),
("prompt", (None, prompt)),
("seconds", (None, "5")),
("size", (None, "1280x720")),
("first_frame", ("first-frame.png", image_1, "image/png")),
("last_frame", ("last-frame.png", image_2, "image/png")),
]
response = requests.post(
"https://api.cometapi.com/v1/videos",
headers={"Authorization": "Bearer " + os.environ["COMETAPI_KEY"]},
files=fields,
timeout=120,
)
response.raise_for_status()
print(response.json())
- lang: JavaScript
label: First and last frames from files
source: >
import { readFile } from "node:fs/promises";
const image_1 = await readFile("/path/to/first-frame.png");
const image_2 = await readFile("/path/to/last-frame.png");
const form = new FormData();
form.append("model", "flux-3");
form.append("prompt", "Move smoothly from the opening to the closing
scene.");
form.append("seconds", "5");
form.append("size", "1280x720");
form.append(
"first_frame",
new Blob([image_1], { type: "image/png" }),
"first-frame.png",
);
form.append(
"last_frame",
new Blob([image_2], { type: "image/png" }),
"last-frame.png",
);
const response = await fetch("https://api.cometapi.com/v1/videos", {
method: "POST",
headers: { Authorization: "Bearer " + process.env.COMETAPI_KEY },
body: form,
});
if (!response.ok) {
throw new Error(await response.text());
}
console.log(await response.json());
- lang: Shell
label: Opening frame from a file
source: |-
curl \
https://api.cometapi.com/v1/videos \
-H "Authorization: Bearer $COMETAPI_KEY" \
--form-string 'model=flux-3' \
--form-string 'prompt=Animate the opening frame with a slow camera move and natural motion.' \
--form-string 'seconds=5' \
--form-string 'size=1280x720' \
--form 'first_frame=@/path/to/reference.png;type=image/png'
- lang: Python
label: Opening frame from a file
source: |
import os
import requests
prompt = (
"Animate the opening frame with a slow camera move and "
"natural motion."
)
with open("/path/to/reference.png", "rb") as reference:
response = requests.post(
"https://api.cometapi.com/v1/videos",
headers={
"Authorization": "Bearer "
+ os.environ["COMETAPI_KEY"]
},
data={
"model": "flux-3",
"prompt": prompt,
"seconds": "5",
"size": "1280x720",
},
files={
"first_frame": (
"reference.png",
reference,
"image/png",
)
},
timeout=120,
)
response.raise_for_status()
print(response.json())
- lang: JavaScript
label: Opening frame from a file
source: |
import { readFile } from "node:fs/promises";
const reference = await readFile("/path/to/reference.png");
const form = new FormData();
form.append("model", "flux-3");
form.append(
"prompt",
"Animate the opening frame with a slow camera move and natural motion.",
);
form.append("seconds", "5");
form.append("size", "1280x720");
form.append(
"first_frame",
new Blob([reference], { type: "image/png" }),
"reference.png",
);
const response = await fetch("https://api.cometapi.com/v1/videos", {
method: "POST",
headers: { Authorization: `Bearer ${process.env.COMETAPI_KEY}` },
body: form,
});
if (!response.ok) {
throw new Error(await response.text());
}
console.log(await response.json());
- lang: Shell
label: Ordered keyframes from image URLs
source: |-
curl \
https://api.cometapi.com/v1/videos \
-H "Authorization: Bearer $COMETAPI_KEY" \
--form-string 'model=flux-3' \
--form-string 'prompt=Move smoothly through the supplied reference images in their given order. Preserve recognizable subjects and visual details as the scene changes. Keep the camera movement gentle.' \
--form-string 'seconds=5' \
--form-string 'size=1280x720' \
--form-string 'input_reference=https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png' \
--form-string 'input_reference=https://apidoc.cometapi.com/images/image/gemini/6100640_569420.png'
- lang: Python
label: Ordered keyframes from image URLs
source: |
import os
import requests
fields = [
("model", (None, "flux-3")),
("prompt", (None, "Move smoothly through the supplied reference images in their given order. Preserve recognizable subjects and visual details as the scene changes. Keep the camera movement gentle.")),
("seconds", (None, "5")),
("size", (None, "1280x720")),
("input_reference", (None, "https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png")),
("input_reference", (None, "https://apidoc.cometapi.com/images/image/gemini/6100640_569420.png")),
]
response = requests.post(
"https://api.cometapi.com/v1/videos",
headers={"Authorization": "Bearer " + os.environ["COMETAPI_KEY"]},
files=fields,
timeout=120,
)
response.raise_for_status()
print(response.json())
- lang: JavaScript
label: Ordered keyframes from image URLs
source: >
const form = new FormData();
form.append("model", "flux-3");
form.append("prompt", "Move smoothly through the supplied reference
images in their given order. Preserve recognizable subjects and
visual details as the scene changes. Keep the camera movement
gentle.");
form.append("seconds", "5");
form.append("size", "1280x720");
form.append("input_reference",
"https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png");
form.append("input_reference",
"https://apidoc.cometapi.com/images/image/gemini/6100640_569420.png");
const response = await fetch("https://api.cometapi.com/v1/videos", {
method: "POST",
headers: { Authorization: `Bearer ${process.env.COMETAPI_KEY}` },
body: form,
});
if (!response.ok) {
throw new Error(await response.text());
}
console.log(await response.json());
components:
schemas:
Flux3CreateRequest:
type: object
description: >-
Flux 3 multipart request. Choose text, input_reference URLs, or
first_frame and last_frame inputs. Keep reference-image and frame inputs
separate.
required:
- model
- prompt
not:
anyOf:
- required:
- input_reference
- first_frame
- required:
- input_reference
- last_frame
properties:
model:
type: string
const: flux-3
default: flux-3
description: Model ID for this route. Use flux-3.
prompt:
type: string
minLength: 1
description: >-
Text that describes the scene, motion, camera behavior, and visual
details that the video should preserve.
default: >-
A paper boat glides across a still pond while the camera moves
forward.
seconds:
type: integer
minimum: 5
maximum: 20
description: >-
Requested clip duration in whole seconds. Set seconds explicitly to
an integer from 5 through 20.
size:
type: string
enum:
- 1280x720
- 1920x1080
default: 1280x720
description: >-
Resolution preset in exact WxH form. Use 1280x720 for the 720p tier
or 1920x1080 for the 1080p tier. Encoded dimensions can be
codec-aligned; image-to-video framing can follow the reference
image.
input_reference:
description: >-
Reference image URLs. Send one publicly accessible HTTPS image URL
per input_reference field. For multiple images, repeat the field in
keyframe order, up to 10 URLs. This field accepts URLs only. Keep it
separate from first_frame and last_frame.
anyOf:
- type: string
format: uri
pattern: ^https://
- type: array
minItems: 1
maxItems: 10
items:
type: string
format: uri
pattern: ^https://
first_frame:
description: >-
Opening-frame image as one publicly accessible HTTPS URL or one PNG
or JPEG file. Keep each file at or below 20 MiB (20 × 1024 × 1024
bytes). Send the field once, using either a URL or a file. Pair with
last_frame to set both boundaries. Keep frame inputs separate from
input_reference.
anyOf:
- type: string
format: uri
pattern: ^https://
- type: string
format: binary
last_frame:
description: >-
Closing-frame image as one publicly accessible HTTPS URL or one PNG
or JPEG file. Keep each file at or below 20 MiB (20 × 1024 × 1024
bytes). Send the field once, using either a URL or a file. To
specify the closing image through this endpoint, send last_frame
together with first_frame. Keep frame inputs separate from
input_reference.
anyOf:
- type: string
format: uri
pattern: ^https://
- type: string
format: binary
additionalProperties: false
dependentRequired:
last_frame:
- first_frame
Flux3VideoTask:
type: object
required:
- id
- object
- model
- status
- progress
- created_at
properties:
id:
type: string
description: Task ID. Use this value as task_id in retrieve and content requests.
example:
task_id:
type: string
description: >-
Compatibility alias for id. This field can be omitted from retrieve
responses.
example:
object:
type: string
const: video
description: Object type for the asynchronous video task.
model:
type: string
const: flux-3
description: Model ID that the task uses.
status:
type: string
enum:
- queued
- in_progress
- completed
- failed
description: Task lifecycle status. Poll until the value is completed or failed.
progress:
type: integer
minimum: 0
maximum: 100
description: Task progress as a coarse percentage.
created_at:
type: integer
format: int64
description: Task creation time as a Unix timestamp in seconds.
completed_at:
type: integer
format: int64
description: >-
Task completion time as a Unix timestamp in seconds when the task
provides one.
expires_at:
type: integer
format: int64
description: >-
Result expiration time as a Unix timestamp in seconds when the task
provides one.
video_url:
type: string
format: uri
description: Video delivery URL. This field appears on completed tasks.
example: https://media.example.com/flux-3-result.mp4
error:
type: object
description: Failure details. This field appears when the task fails.
properties:
message:
type: string
description: Human-readable failure description.
code:
type: string
description: Failure code when the task provides one.
additionalProperties: true
additionalProperties: true
securitySchemes:
bearerAuth:
type: http
scheme: bearer
description: Bearer authentication. Use your CometAPI API key.
````