openapi: 3.1.0 info: title: PixAI API - OpenAPI 2.0 description: | Users own the copyrights to images generated through the API. PixAI does not claim any copyright over user-generated works. However, users must ensure that generated content does not infringe any third-party rights and complies with applicable local laws and regulations. PixAI's official models (e.g. Haruka, Tsubaki) are copyrighted by PixAI. We grant approved users the right to use these models for commercial activities, but reselling or redistributing the model files themselves is prohibited. For details, please refer to: termsOfService: https://pixai.art/terms license: name: Apache 2.0 url: http://www.apache.org/licenses/LICENSE-2.0.html version: 2.0.0 servers: - url: https://api.pixai.art/v2 tags: - name: image description: Image generation and editing endpoints - name: video description: Video generation endpoints - name: task description: Read a task and its results - name: media description: Upload input files and resolve media IDs to URLs - name: model description: Find models, LoRAs, sizes and color palettes paths: /image/create: post: tags: - image summary: Create an image generation task description: | Start an asynchronous image generation task with the simplified v2 API. The response returns the new task's ID. The v2 API provides a streamlined interface with user-friendly parameters such as `aspectRatio`, `mode`, and named `style` presets, replacing the raw pixel dimensions and internal parameter names of v1. operationId: createImage x-codeSamples: - id: curl lang: bash label: cURL source: | curl -X POST https://api.pixai.art/v2/image/create \ -H "Authorization: Bearer $PIXAI_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "modelVersionId": "1983308862240288769", "prompt": "1girl, green hair, solo, standing", "aspectRatio": "9:16", "mode": "standard", "batchSize": 4, "seed": 12345 }' - id: js lang: javascript label: JavaScript (fetch) source: | const response = await fetch("https://api.pixai.art/v2/image/create", { method: "POST", headers: { "Authorization": `Bearer ${PIXAI_API_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ modelVersionId: "1983308862240288769", prompt: "1girl, green hair, solo, standing", aspectRatio: "9:16", mode: "standard", batchSize: 4, seed: 12345, }), }); const task = await response.json(); console.log(task); - id: python lang: python label: Python (requests) source: | import requests response = requests.post( "https://api.pixai.art/v2/image/create", headers={ "Authorization": f"Bearer {PIXAI_API_KEY}", "Content-Type": "application/json", }, json={ "modelVersionId": "1983308862240288769", "prompt": "1girl, green hair, solo, standing", "aspectRatio": "9:16", "mode": "standard", "batchSize": 4, "seed": 12345, }, ) task = response.json() print(task) requestBody: required: true content: application/json: schema: $ref: "#/components/schemas/CreateImageRequest" example: modelVersionId: "1983308862240288769" prompt: "1girl, green hair, solo, standing" aspectRatio: "9:16" mode: "standard" batchSize: 4 seed: 12345 responses: "201": description: | Image generation task created successfully. Example Response: ```json { "id": "1943547603237249965", "status": "waiting", "createdAt": "2025-11-07T14:54:35.000Z", "updatedAt": "2025-11-07T14:54:35.000Z", "outputs": { "mediaIds": [], "mediaUrls": [] } } ``` content: application/json: schema: $ref: "#/components/schemas/Task" "400": description: | The request was rejected before a task was created. `code` says why: - `BAD_REQUEST` with `data.issues`: the body failed validation (each issue names the field in `path`), or a field is not supported by this model (`message` says which). - `INVALID_ARGUMENT`: a field is not allowed for this model family, e.g. `mode` on an SDXL model or `sampling` on Tsubaki. Only `mode` and `sampling` are checked: `mode` works only on Tsubaki models, `sampling` only on SDXL and SD v1. Remove whichever the model does not accept. - `IMAGE_SIZE_NOT_ALLOWED`: `width`/`height` fall outside the model's size ranges (`data.reason`). Pick a size from `GET /v2/generation-model/{modelVersionId}/size-config`, or send `aspectRatio`/`size` instead. - `INPUT_IMAGE_TOO_LARGE`: a `contextImages` entry is 4000 px or larger on a side. content: application/json: schema: $ref: "#/components/schemas/Error" example: defined: false code: BAD_REQUEST status: 400 message: Input validation failed data: issues: - code: custom path: [] message: Either modelId or modelVersionId must be provided "401": description: Authentication required. Ensure a valid Bearer token is provided. content: application/json: schema: $ref: "#/components/schemas/Error" "403": description: | The account cannot run this task right now. `code` says why: - `INSUFFICIENT_BALANCE`: not enough credits for the task. - `WAITING_NUM_EXCEEDED`: too many of your tasks are still `waiting`. Wait for some to start before submitting more (see [Limits](/en/docs/quick-start/limits)). - `LORA_NUM_EXCEEDED`: more `loras` than your plan allows. content: application/json: schema: $ref: "#/components/schemas/Error" "404": description: | `NOT_FOUND`: no model version has the given `modelVersionId`. `FILE_NOT_FOUND`: a `contextImages` media ID does not exist. content: application/json: schema: $ref: "#/components/schemas/Error" "422": description: | - `NSFW_PROMPT_REJECTED`: the prompt was rejected by content moderation. Rephrase it; `data.logId` identifies the decision if you contact support. - `INVALID_INPUT`: a value is not available for this model or plan, e.g. a `mode` the model does not offer. `message` names the value. content: application/json: schema: $ref: "#/components/schemas/Error" "500": description: Task creation failed due to an internal server error. Retry with exponential backoff. content: application/json: schema: $ref: "#/components/schemas/Error" security: - bearerAuth: [] /frame-video/create: post: tags: - video summary: Create an image-to-video task description: | Animates an image into a video with PixAI's Tsubaki video model. The video starts on `firstFrameMediaId` and, when you send `lastFrameMediaId`, ends on it. The output keeps the first frame's aspect ratio. Frames are image media IDs: upload your own with `POST /v2/media/image/upload-url`, or reuse an image you generated. Media owned by another user is rejected. Image-to-video needs a PixAI membership. Credits are charged when the task is created. Poll `GET /v2/task/{id}` for the result; a finished task's video is `outputs.videoMediaId`. operationId: createFrameVideo x-codeSamples: - id: curl lang: bash label: cURL source: | curl -X POST https://api.pixai.art/v2/frame-video/create \ -H "Authorization: Bearer $PIXAI_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "tbkv1.0.1", "firstFrameMediaId": "1943547603237249965", "prompt": "she turns around and smiles, hair blowing in the wind", "duration": 5 }' requestBody: required: true content: application/json: schema: type: object properties: model: enum: - tbkv1.0.1 - tbkv1.0 type: string description: The video model. `tbkv1.0.1` is the latest Tsubaki video model. firstFrameMediaId: type: string description: Image media ID of the first frame. lastFrameMediaId: type: string description: Image media ID of the last frame. Send the first frame again to make a loop. prompt: default: null type: - string - "null" description: What should happen in the video. duration: type: integer minimum: 5 maximum: 15 default: 5 description: Length in whole seconds. mode: enum: - basic - professional type: string default: basic description: "`professional` adds a 1080p upscale on `tbkv1.0.1` and costs more. It has no effect on `tbkv1.0`." generateAudio: type: boolean default: true description: Whether to generate a soundtrack. audioLanguage: default: null enum: - none - chinese - english - japanese - korean - null type: - string - "null" description: Language of speech in the soundtrack. `none` for no speech. required: - model - firstFrameMediaId additionalProperties: false example: model: tbkv1.0.1 firstFrameMediaId: "1943547603237249965" prompt: she turns around and smiles, hair blowing in the wind duration: 5 responses: "201": description: The task was created. Poll `GET /v2/task/{id}` with its `id`. content: application/json: schema: $ref: "#/components/schemas/Task" "400": description: | - `INVALID_REFERENCE_MEDIA`: a frame is missing, is not an image, or belongs to another user. `data.code` is `invalid_input` or `media_not_owned`, and `data.mediaId` names the frame. - `BAD_REQUEST` with `data.issues`: the body failed validation. content: application/json: schema: $ref: "#/components/schemas/Error" "401": $ref: "#/components/responses/Unauthorized" "403": $ref: "#/components/responses/CannotRun" "500": $ref: "#/components/responses/CreationFailed" security: - bearerAuth: [] /reference-video/create: post: tags: - video summary: Create a reference video task description: | Generates a video guided by reference images, and depending on the model, reference videos and audio. Each model accepts different references: - `tbkv1.0.1`, `tbkv1.0` (Tsubaki): 1–6 images, no videos, up to 3 audio clips, `duration` 5–15. - `v4.0`, `v4.0.1` (Seedance 2, Seedance 2 Fast): up to 9 images, up to 3 videos, up to 3 audio clips, `duration` 4–15. - `v4.0.3` (Seedance 2.5): up to 30 images, up to 10 videos, up to 10 audio clips, `duration` 4–30. - `v3.3` (Vidu): 1–7 images, no videos or audio, `duration` 3–16. Every model needs at least one image or video, except `v4.0.3`, which also accepts audio alone. Reference files must meet these limits: - Video: MP4 or MOV, H.264 or HEVC, 24–60 fps, 2–15 s per clip and 15 s in total (30 s on `v4.0.3`), up to 50 MB (200 MB on `v4.0.3`). - Audio: WAV or MP3, 2–15 s per clip and 15 s in total (30 s on `v4.0.3`), under 15 MiB. Check a file with `GET /v2/media/{mediaId}/info` before sending it. On `v4.0` and `v4.0.1`, counts and clip lengths are checked after the task is charged: a task that breaks them fails instead of being refused. Credits are charged when the task is created. Poll `GET /v2/task/{id}` for the result; a finished task's video is `outputs.videoMediaId`. operationId: createReferenceVideo x-codeSamples: - id: curl lang: bash label: cURL source: | curl -X POST https://api.pixai.art/v2/reference-video/create \ -H "Authorization: Bearer $PIXAI_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "tbkv1.0.1", "prompt": "the girl walks through the city at night", "referenceImageMediaIds": ["1943547603237249965"], "ratio": "9:16", "duration": 5 }' requestBody: required: true content: application/json: schema: type: object properties: model: type: string minLength: 1 enum: - tbkv1.0.1 - tbkv1.0 - v4.0 - v4.0.1 - v4.0.3 - v3.3 description: The video model. See the table above for what each accepts. An unknown value answers `403 MODEL_VERSION_NOT_EXISTS`. prompt: default: null type: - string - "null" description: What should happen in the video. Describe how to use the references. referenceImageMediaIds: type: array items: type: string default: [] description: Image media IDs. referenceVideoMediaIds: type: array items: type: string default: [] description: Video media IDs. Not accepted by Tsubaki or Vidu. referenceAudioMediaIds: type: array items: type: string default: [] description: Audio media IDs, uploaded with `POST /v2/media/audio/upload-url`. Not accepted by Vidu. duration: type: integer default: 5 description: Length in whole seconds. The allowed range depends on `model`. ratio: enum: - adaptive - 1:1 - 2:3 - 3:2 - 3:4 - 4:3 - 9:16 - 16:9 - 21:9 type: string description: | Output aspect ratio. Omitted or `adaptive` follows the references. Seedance models do not take `2:3` or `3:2`, and Vidu ignores this field. mode: enum: - basic - professional type: string default: basic description: "`professional` adds an upscale (1080p on `v4.0`, 720p on `v4.0.1`) and costs more. Not accepted by `v4.0.3`." generateAudio: type: boolean default: true description: Whether to generate a soundtrack. audioLanguage: default: null enum: - none - chinese - english - japanese - korean - null type: - string - "null" description: Language of speech in the soundtrack. `none` for no speech. required: - model example: model: tbkv1.0.1 prompt: the girl walks through the city at night referenceImageMediaIds: - "1943547603237249965" ratio: 9:16 duration: 5 responses: "201": description: The task was created. Poll `GET /v2/task/{id}` with its `id`. content: application/json: schema: $ref: "#/components/schemas/Task" "400": description: | - `BAD_REQUEST` with `data.issues`: the body failed validation, e.g. too many references for `v4.0.3`. - `VALIDATION_ERROR`: the combination is not valid for this model, e.g. `mode: professional` on `v4.0.3`, or a video or audio reference for Vidu. `message` says which. content: application/json: schema: $ref: "#/components/schemas/Error" "401": $ref: "#/components/responses/Unauthorized" "403": $ref: "#/components/responses/CannotRun" "422": description: | `INVALID_INPUT`: a reference file was refused before the task was charged. `message` reads `Invalid reference media: `, sometimes followed by the media ID in parentheses. Reasons include `too_many_reference_images`, `reference_video_not_supported` and `unsupported_reference_video_format`. content: application/json: schema: $ref: "#/components/schemas/Error" "500": $ref: "#/components/responses/CreationFailed" security: - bearerAuth: [] /gen-task/inpaint: post: tags: - image summary: Create an inpaint, outpaint or redraw task description: | Edits an existing image: - **Inpaint** (`fillType: inpaint`): redraws the masked area. - **Outpaint** (`fillType: outpaint`): extends the image beyond its edges. Place the original on a larger white canvas and send that as `baseImageMediaId`; the mask marks the new area. - **Redraw** (`fillType: inpaint`, no mask): redraws the whole image, keeping it closer to the original as `strength` goes down. A masked edit takes two extra images, always sent together: - `maskMediaId`: same size as the base image, white where to redraw and black where to keep. - `maskedImageMediaId`: the base image with the white area of the mask painted gray (`#808080`). The output has the size of the masked image when a mask is sent, otherwise of the base image. Images are media IDs; see `POST /v2/media/image/upload-url`. Use `modelPreset: "v1"`. `GET /v2/gen-task/inpaint/constraints` lists the size limits and `sdeditPreset` names of a preset. Credits are charged when the task is created; the result is `outputs.mediaIds[0]` of `GET /v2/task/{id}`. operationId: createInpaint x-codeSamples: - id: curl lang: bash label: cURL source: | curl -X POST https://api.pixai.art/v2/gen-task/inpaint \ -H "Authorization: Bearer $PIXAI_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "modelPreset": "v1", "fillType": "inpaint", "baseImageMediaId": "1943547603237249965", "maskMediaId": "1943547603237249966", "prompt": "red scarf" }' requestBody: required: true content: application/json: schema: type: object properties: modelPreset: type: string minLength: 1 description: The inpaint model. Use `v1`. example: v1 fillType: enum: - inpaint - outpaint type: string baseImageMediaId: type: string description: Image media ID of the image to edit. maskMediaId: type: string description: Image media ID of the mask, white where to redraw and black where to keep. Send it with `maskedImageMediaId`, or omit both to redraw the whole image. maskedImageMediaId: type: string description: Image media ID of the base image with the mask's white area painted `#808080`. Send it with `maskMediaId`. prompt: type: string maxLength: 10000 default: "" description: What to draw in the edited area. strength: type: number minimum: 0 maximum: 1 multipleOf: 0.01 description: How far the result may depart from the original. Lower keeps it closer. seed: type: integer description: Random seed. Omit for a random one. promptHelper: enum: - enable - disable - low - medium type: string default: enable description: How much PixAI may expand the prompt. sdeditPreset: type: string minLength: 1 maxLength: 64 description: A redraw preset name from `GET /v2/gen-task/inpaint/constraints`. Sending one to a preset without any answers `SDEDIT_NOT_AVAILABLE`. sdedit: type: object properties: spectralScale: type: number minimum: 0 maximum: 1 hfsBlurSigma: type: number minimum: 0 maximum: 32 hfsEnd: type: number minimum: 0 maximum: 1 colorCorrection: type: boolean description: Match the colors of the original. description: Fine-tuning for redraws. With `sdeditPreset`, only `colorCorrection` is taken from here. required: - baseImageMediaId - fillType - modelPreset example: modelPreset: v1 fillType: inpaint baseImageMediaId: "1943547603237249965" maskMediaId: "1943547603237249966" prompt: red scarf responses: "201": description: The task was created. Poll `GET /v2/task/{id}` with its `id`. content: application/json: schema: $ref: "#/components/schemas/Task" "400": description: | - `INVALID_MODEL_PRESET`: unknown `modelPreset`. - `INVALID_MEDIA`: an image media ID does not exist or is not an image. - `INVALID_SDEDIT_PRESET`: unknown `sdeditPreset`. - `SDEDIT_NOT_AVAILABLE`: this `modelPreset` takes no `sdeditPreset` or `sdedit`. - `BAD_REQUEST` with `data.issues`: the body failed validation. content: application/json: schema: $ref: "#/components/schemas/Error" "401": $ref: "#/components/responses/Unauthorized" "403": $ref: "#/components/responses/CannotRun" "500": $ref: "#/components/responses/CreationFailed" security: - bearerAuth: [] /gen-task/inpaint/constraints: get: tags: - image summary: Get the limits of an inpaint preset description: Returns the size limits and redraw presets of an inpaint `modelPreset`. An unknown preset answers `null` fields rather than an error. No authentication is needed. operationId: getInpaintConstraints parameters: - name: modelPreset in: query required: true schema: type: string minLength: 1 example: v1 responses: "200": description: The preset's limits. content: application/json: schema: type: object properties: modelPreset: type: string constraints: type: - object - "null" description: Size limits for the image, or `null` if the preset sets none. properties: minWidth: type: integer maxWidth: type: integer minHeight: type: integer maxHeight: type: integer maxPixels: type: integer description: Maximum width × height. dimensionStep: type: integer description: Width and height must be multiples of this. sdeditPresets: type: - object - "null" description: Redraw presets by name, for `sdeditPreset`. additionalProperties: type: object defaultSdeditPreset: type: - string - "null" security: - {} - bearerAuth: [] /task/{id}: get: tags: - task summary: Get a task description: | Reads a task created by any v2 endpoint. Poll it every few seconds until `status` is `completed`, `failed` or `cancelled`. Where the result is: - Images (`taskKind` `t2i`, `i2i`, `inpaint`): `outputs.mediaIds` and `outputs.mediaUrls`. - Videos (`taskKind` `i2v-inhouse`, `reference-video`): `outputs.videoMediaId`. Resolve it with `GET /v2/media/batch?ids=` and download `fileUrl`. `outputs.mediaIds` then holds only the poster image. Generated media is not kept forever. Download results soon after they finish. operationId: getTask x-codeSamples: - id: curl lang: bash label: cURL source: | curl https://api.pixai.art/v2/task/1943547603237249965 \ -H "Authorization: Bearer $PIXAI_API_KEY" parameters: - name: id in: path required: true schema: type: string description: The task ID. - name: with_parameters in: query schema: type: string enum: - "true" description: Send `true` to include the stored task parameters in `parameters`. responses: "200": description: The task. content: application/json: schema: $ref: "#/components/schemas/Task" "401": $ref: "#/components/responses/Unauthorized" "404": description: "`NOT_FOUND`: no task with this ID is visible to you. A request without an API key gets this for every task." content: application/json: schema: $ref: "#/components/schemas/Error" security: - bearerAuth: [] /media/image/upload-url: post: tags: - media summary: Get an upload URL for an image description: | Step 1 of uploading an image you want to use as an input, for example a `contextImages` entry, a video frame or an inpaint base image. 1. Call this endpoint to get `uploadUrl` and `externalId`. 2. Send the raw image bytes to `uploadUrl` with an HTTP `PUT` and the file's real `Content-Type` (for example `image/png`). Do not use a multipart form. 3. Call `POST /v2/media/image/register` with the `externalId`. Its `id` is the media ID you pass to other endpoints. operationId: getImageUploadUrl x-codeSamples: - id: curl lang: bash label: cURL source: | # 1. Get an upload URL UPLOAD=$(curl -s -X POST https://api.pixai.art/v2/media/image/upload-url \ -H "Authorization: Bearer $PIXAI_API_KEY" \ -H "Content-Type: application/json" -d '{}') URL=$(echo "$UPLOAD" | jq -r .uploadUrl) EXTERNAL_ID=$(echo "$UPLOAD" | jq -r .externalId) # 2. Upload the bytes curl -X PUT "$URL" -H "Content-Type: image/png" --data-binary @input.png # 3. Register it and read the media ID curl -s -X POST https://api.pixai.art/v2/media/image/register \ -H "Authorization: Bearer $PIXAI_API_KEY" \ -H "Content-Type: application/json" \ -d "{\"externalId\": \"$EXTERNAL_ID\"}" | jq -r .id requestBody: required: true content: application/json: schema: type: object properties: isEphemeral: type: boolean default: false description: Mark the file as a short-lived generation input. Ephemeral files are private, and their URLs are signed and expire. example: {} responses: "200": description: The upload target. content: application/json: schema: $ref: "#/components/schemas/UploadUrl" "401": $ref: "#/components/responses/Unauthorized" security: - bearerAuth: [] /media/image/register: post: tags: - media summary: Register an uploaded image description: | Step 3 of uploading an image: turns the uploaded file into a media item. Call it after the `PUT` to `uploadUrl` has succeeded. The returned `id` is the media ID to pass to other endpoints. operationId: registerImage requestBody: required: true content: application/json: schema: type: object properties: externalId: type: string format: uuid description: The `externalId` returned by `POST /v2/media/image/upload-url`. isEphemeral: type: boolean description: Send the same value you sent to `upload-url`. required: - externalId example: externalId: 3f2b8c1e-6a4d-4f6e-9b1a-2c7d5e8f9a01 responses: "200": description: The registered image. `width` and `height` can be `0` until processing finishes a few seconds later; read them from `GET /v2/media/batch` if you need them. content: application/json: schema: $ref: "#/components/schemas/Media" "401": $ref: "#/components/responses/Unauthorized" security: - bearerAuth: [] /media/audio/upload-url: post: tags: - media summary: Get an upload URL for an audio file description: | Step 1 of uploading an audio clip for `referenceAudioMediaIds` of `POST /v2/reference-video/create`. The flow is the same as for images: `PUT` the bytes to `uploadUrl` with the file's `Content-Type` (`audio/mpeg` or `audio/wav`), then call `POST /v2/media/audio/register`. Only WAV and MP3 are accepted, under 15 MiB per clip. Duration limits depend on the video model; they are checked when the video task is created, not at upload. operationId: getAudioUploadUrl requestBody: required: true content: application/json: schema: type: object properties: isEphemeral: type: boolean default: false description: Mark the file as a short-lived generation input. Ephemeral files are private, and their URLs are signed and expire. example: {} responses: "200": description: The upload target. content: application/json: schema: $ref: "#/components/schemas/UploadUrl" "401": $ref: "#/components/responses/Unauthorized" security: - bearerAuth: [] /media/audio/register: post: tags: - media summary: Register an uploaded audio file description: Step 3 of uploading an audio clip. The returned `id` is the media ID to pass in `referenceAudioMediaIds`. operationId: registerAudio requestBody: required: true content: application/json: schema: type: object properties: externalId: type: string minLength: 1 description: The `externalId` returned by `POST /v2/media/audio/upload-url`. required: - externalId example: externalId: 3f2b8c1e-6a4d-4f6e-9b1a-2c7d5e8f9a01 responses: "200": description: The registered audio file. content: application/json: schema: type: object required: - id properties: id: type: string description: The media ID. fileUrl: type: - string - "null" description: URL of the file. duration: type: - number - "null" description: Length in seconds. size: type: - number - "null" description: Size in bytes. codec: type: - string - "null" "401": $ref: "#/components/responses/Unauthorized" security: - bearerAuth: [] /media/{mediaId}/info: get: tags: - media summary: Inspect a video or audio file description: | Returns the duration, size, containers and codecs of a video or audio media item, so you can check a reference file against the model's limits before creating a reference video task. operationId: getMediaInfo parameters: - name: mediaId in: path required: true schema: type: string description: A video or audio media ID. responses: "200": description: The media metadata. content: application/json: schema: type: object properties: mediaId: type: string durationSeconds: type: - number - "null" description: Length in seconds. sizeBytes: type: - string - "null" description: Size in bytes, as a decimal string. containers: type: array items: type: string description: Container formats, lowercase, e.g. `mp3` or `mov`. videoStreams: type: array description: One entry per video stream. items: type: object properties: codec: type: string width: type: integer height: type: integer averageFrameRate: type: number audioStreams: type: array description: One entry per audio stream. items: type: object properties: codec: type: string "400": description: "`NOT_VIDEO_OR_AUDIO`: the media is an image." content: application/json: schema: $ref: "#/components/schemas/Error" "401": $ref: "#/components/responses/Unauthorized" "404": description: "`MEDIA_NOT_FOUND`: no media has this ID." content: application/json: schema: $ref: "#/components/schemas/Error" security: - bearerAuth: [] /media/batch: get: tags: - media summary: Get media URLs by ID description: | Resolves up to 100 media IDs to their URLs, for example the `outputs.mediaIds` or `outputs.videoMediaId` of a finished task. No authentication is needed for public media. IDs that do not exist are left out, and the order of the result is not guaranteed, so match items by `id`. operationId: batchGetMedia parameters: - name: ids in: query required: true schema: anyOf: - type: string - type: array items: type: string description: Media IDs to resolve (1–100). Accepts comma-separated (`?ids=1,2,3`), bracket notation (`?ids[]=1&ids[]=2`), or repeated (`?ids=1&ids=2`). type: string description: Comma-separated media IDs, 1 to 100, e.g. `?ids=1,2,3`. Repeating the parameter (`?ids=1&ids=2`) also works. responses: "200": description: The media items found. content: application/json: schema: type: array items: $ref: "#/components/schemas/Media" security: - {} - bearerAuth: [] /generation-model/search: get: tags: - model summary: Search models and LoRAs description: | Searches the [PixAI Model Market](https://pixai.art/model). Use it to find a base model for `modelVersionId`, or LoRAs for `loras` of `POST /v2/image/create`. To use a result, pass its `latestAvailableVersionId`, not its `id`: generation endpoints take **version** IDs. A LoRA's default trigger words are in `latestAvailableVersion.extra.triggerWords`. To find LoRAs that work with a base model, send `usageType=LORA` and the base model's type in `loraBaseModelTypes` (the `type` of a model item, e.g. `SDXL_MODEL`). Results depend on who is asking: send your API key to get the same content rating as your account. Anonymous requests get stricter filtering. operationId: searchGenerationModels parameters: - name: usageType in: query required: true schema: anyOf: - const: ALL - const: MODEL - const: LORA type: string enum: - ALL - MODEL - LORA description: "`MODEL` for base models, `LORA` for LoRAs, `ALL` for both." - name: keyword in: query schema: type: string minLength: 1 maxLength: 256 description: Free-text query. Results rank by relevance and popularity; without a keyword, by popularity. - name: loraBaseModelTypes in: query schema: type: string description: "With `usageType=LORA`: comma-separated base model types the LoRA must work with, e.g. `SDXL_MODEL`." - name: baseModelIds in: query schema: type: string description: "With `usageType=LORA`: comma-separated base model IDs the LoRA must work with." - name: source in: query schema: enum: - ANY - PIXAI - EXTERNAL type: string default: ANY description: "With `usageType=LORA`: `PIXAI` for LoRAs trained on PixAI, `EXTERNAL` for imported ones." - name: baseModelType in: query schema: type: string description: "With `usageType=MODEL`: comma-separated model types, e.g. `SDXL_MODEL,MMDIT26B_MODEL`." - name: permittedUse in: query schema: enum: - COMMERCIAL - PERSONAL type: string description: Only models whose license allows this use. - name: postedWithin in: query schema: enum: - today - this-week - this-month type: string - name: authorId in: query schema: type: string description: Only models by this user. - name: size in: query schema: type: integer maximum: 50 exclusiveMinimum: 0 default: 24 minimum: 1 description: Page size. - name: offset in: query schema: type: integer minimum: 0 maximum: 10000 default: 0 description: Number of results to skip. Pages can overlap slightly; de-duplicate by `id`. responses: "200": description: One page of results. content: application/json: schema: type: object required: - data - hasMore properties: data: type: array items: $ref: "#/components/schemas/ModelListItem" hasMore: type: boolean description: Whether another page may follow. security: - {} - bearerAuth: [] /generation-model/{id}: get: tags: - model summary: Get a model or LoRA description: Reads one model by its **model** ID (the `id` of a search result). operationId: getGenerationModel parameters: - name: id in: path required: true schema: type: string description: The model ID. responses: "200": description: The model. content: application/json: schema: allOf: - $ref: "#/components/schemas/ModelListItem" - type: object properties: triggerWords: type: - string - "null" description: Default trigger words of the latest version. "404": $ref: "#/components/responses/ModelNotFound" security: - {} - bearerAuth: [] /generation-model/{modelId}/versions: get: tags: - model summary: List the versions of a model description: Lists a model's versions, newest first. Pass a version's `id` as `modelVersionId` or `loras[].modelId` to pin an older version. operationId: listGenerationModelVersions parameters: - name: modelId in: path required: true schema: type: string description: The model ID. - name: baseModelType in: query schema: enum: - ANY_LORA - ANY_MODEL - ANY_NON_SDXL_LORA - ANY_NON_USER_LORA - ANY_SDXL_LORA - ANY_USER_LORA - ANY_VAE - CHAT - DIT5A_MODEL_INTERNAL - DIT7A_LORA - DIT7A_MODEL - DIT7A_VAE - DIT7B_LORA - DIT7B_MODEL - DIT7B_VAE - DIT7C_LORA - DIT7C_MODEL - DIT7C_VAE - DIT7D_LORA - DIT7D_MODEL - DIT7D_VAE - DIT7_LORA - DIT7_MODEL - DIT7_VAE - DIT9_LORA - DIT9_MODEL - DIT9_VAE - LORA - MMDIT26A_MODEL - MMDIT26B_MODEL - MULTI_LORA - SD3_MEDIUM_MODEL - SDXL_LORA - SDXL_MODEL - SDXL_VAE - SD_V1_MODEL - UNKNOWN_LORA - UNKNOWN_MODEL - USER_DIT26A_MODEL - USER_DIT26B_MODEL - USER_LORA - USER_MULTI_LORA - USER_SDXL_LORA - VAE - VIDEO_MODEL - VIDEO_MODEL_LORA - Z_IMAGE_V1_MODEL type: string description: Only versions for this base model type. responses: "200": description: The versions. content: application/json: schema: type: array items: $ref: "#/components/schemas/ModelVersion" "404": $ref: "#/components/responses/ModelNotFound" security: - {} - bearerAuth: [] /generation-model-version/{id}: get: tags: - model summary: Get a model version description: Reads one model version by its **version** ID, the ID that generation endpoints take. Use it to look up a LoRA's trigger words and base model before generating. operationId: getGenerationModelVersion parameters: - name: id in: path required: true schema: type: string description: The model version ID. responses: "200": description: The version. content: application/json: schema: $ref: "#/components/schemas/ModelVersion" "404": $ref: "#/components/responses/ModelNotFound" security: - {} - bearerAuth: [] /generation-model/{modelVersionId}/size-config: get: tags: - model summary: Get the sizes a model version accepts description: | Lists the sizes a model version accepts for `width`/`height` of `POST /v2/image/create`. Despite the path, the parameter is a **version** ID. A custom `width`/`height` must fall inside one range's `minWidth`–`maxWidth` and `minHeight`–`maxHeight` and be a multiple of its `step`. Presets are ready-made sizes and may lie outside those bounds. Entries whose `accessStatus` is `membership_required` need a higher membership tier than your account has. operationId: getModelSizeConfig parameters: - name: modelVersionId in: path required: true schema: type: string description: The model version ID. responses: "200": description: The size ranges. content: application/json: schema: type: object properties: ranges: type: array items: type: object properties: name: type: string minWidth: type: integer maxWidth: type: integer minHeight: type: integer maxHeight: type: integer step: type: integer description: Width and height must be multiples of this. accessStatus: type: string enum: - available - membership_required presets: type: array items: type: object properties: id: type: string width: type: integer height: type: integer ratioLabel: type: string description: e.g. `9:16`. accessStatus: type: string enum: - available - membership_required "404": $ref: "#/components/responses/ModelNotFound" security: - {} - bearerAuth: [] /color-palettes/presets: get: tags: - model summary: List color palette presets description: | Lists PixAI's official color palettes for `colorPalette` of `POST /v2/image/create`. To use one, send the whole preset object with `"source": "preset"` added. `name` is localized: pass `?lang=ja` (or `ko`, `en`), or an `Accept-Language` header. operationId: listColorPalettePresets responses: "200": description: The presets. content: application/json: schema: type: object properties: palettes: type: array items: allOf: - $ref: "#/components/schemas/ColorPalette" - type: object properties: coverUrl: type: - string - "null" description: Preview image. security: - {} - bearerAuth: [] components: schemas: CreateImageRequest: type: object required: - modelVersionId - prompt additionalProperties: false properties: modelVersionId: type: string minLength: 1 description: | The unique identifier for the model version to use for generation. Preferred replacement for the deprecated `modelId` field. How to find modelVersionId: 1. Open the model page on PixAI. 2. Click the model version you'd like to use. 3. Copy the last path segment of the URL — that's the modelVersionId. Example URL: `https://pixai.art/model/-/` → modelVersionId is ``. Find modelVersionId example: "1983308862240288769" modelId: type: string minLength: 1 x-fold: true x-badge: - type: deprecated info: | Please use `modelVersionId` instead. `modelId` is ambiguous: it actually expects a model **version** ID instead of a model ID. description: | The unique identifier for the model version to use for generation. Taken from the model version's URL on PixAI. How to find modelId: 1. Open the model page on PixAI. 2. Click the model version you'd like to use. 3. Copy the last path segment of the URL — that's the modelId. Example URL: `https://pixai.art/model/-/` → modelId is ``. Find modelId example: "1983308862240288769" prompt: type: string minLength: 1 maxLength: 10000 description: | The main prompt describing the content you want to generate. Up to 10,000 characters on Tsubaki.2 and Tsubaki.3, 4,096 on other models. example: "1girl, green hair, solo, standing" negativePrompt: type: string maxLength: 4096 description: | Text describing elements to avoid in the generated image. example: "nsfw, lowres, bad anatomy" aspectRatio: type: string enum: - "1:1" - "2:3" - "3:2" - "3:4" - "4:3" - "3:5" - "5:3" - "9:16" - "16:9" - "1:3" - "3:1" description: | The aspect ratio of the generated image. Defaults to `1:1` when neither `aspectRatio`/`size` nor `width`/`height` is sent. Cannot be combined with `width` and `height`. Available ratios: `1:1`, `2:3`, `3:2`, `3:4`, `4:3`, `3:5`, `5:3`, `9:16`, `16:9`, `1:3`, `3:1` All ratios are supported across all model types (SDXL, DiT, Tsubaki.2). See [Supported Resolutions](/en/docs/references/supported-resolutions) for the full resolution mapping per size tier. example: "1:1" size: type: string enum: - "1k" - "1.5k" description: | The output resolution tier. Defaults to `1k` when neither `aspectRatio`/`size` nor `width`/`height` is sent. Cannot be combined with `width` and `height`. The longest edge is capped at 1536 px for `1.5k`. - `1k`: ~1 megapixel (e.g. 1024×1024 at 1:1). - `1.5k`: ~2.36 megapixels (e.g. 1536×1536 at 1:1). See [Supported Resolutions](/en/docs/references/supported-resolutions) for the full mapping table. example: "1k" width: type: integer minimum: 1 maximum: 10240 description: | Exact output width in pixels, for sizes the `aspectRatio`/`size` presets do not cover (e.g. Tsubaki.3's larger outputs). Send together with `height` and without `aspectRatio` and `size`. The size must fall within one of the model's size ranges and be a multiple of its `step`. Read them from `GET https://api.pixai.art/v2/generation-model/{modelVersionId}/size-config` (no authentication needed): each entry of `ranges` gives `minWidth`/`maxWidth`, `minHeight`/`maxHeight` and `step`, plus ready-made `presets`. A size outside every range returns `400 IMAGE_SIZE_NOT_ALLOWED`. example: 1024 height: type: integer minimum: 1 maximum: 10240 description: | Exact output height in pixels. Send together with `width`; see `width` for the allowed sizes. example: 1024 mode: type: string x-badge: - type: tsubaki2 info: | Only available for the **Tsubaki.2** and **Tsubaki.3** model families. Rejected with `400 INVALID_ARGUMENT` on other model types. enum: - lite - standard - pro - ultra description: | The inference profile controlling generation quality and speed. - `lite`: Fastest, basic quality. - `standard`: Balanced speed and quality. - `pro`: Higher quality, slower. - `ultra`: Highest quality, slowest. Not every mode is offered on every model or plan; an unavailable mode returns `422 INVALID_INPUT` naming the mode. example: standard style: x-badge: - type: tsubaki2 info: | Only available for the **Tsubaki.2** and **Tsubaki.3** model families. Silently ignored for other model types. description: | The style to apply during image generation. Can be a named preset or a custom style string. oneOf: - type: object additionalProperties: false required: - type - key properties: type: type: string enum: - preset key: type: string enum: - anime-watercolor-impasto - art-crayon - bold-line-soft-color - chibi - classic-japanese - clean-flat-illustration - colored-pencil-grain - cool-warm-oil-contrast - desaturated-cool - flat-anime-pastel - glossy-glass-anime - grey-blocks-lineless - hard-line-monochrome-accent - high-contrast-retro-vivid - holographic - korean-style - lineart-watercolor-airy - low-saturation-expressive-eyes - lucid-dreamy - luminous-retro-impasto - minimal-flat-color - modern-oil-glamour - moe-pastel - oil-watercolor-blend - painterly-doodle - paper-watercolor - pencil-sketch - poster-graffiti - retro-comic-fusion - retro-oil-glamour - retro-tv-anime - semi-realistic - stable-vaporwave - tinted-sketch - vintage-collage description: | A named preset style key. Available presets: - `anime-watercolor-impasto` — Watercolor Glow - `art-crayon` — Art Crayon - `bold-line-soft-color` — Glint & Gloom - `chibi` — Chibi - `classic-japanese` — Classic Anime - `clean-flat-illustration` — Clean Flat-Illustration - `colored-pencil-grain` — Chromatic Pencil - `cool-warm-oil-contrast` — Seasonal Glow - `desaturated-cool` — Cool Softlight - `flat-anime-pastel` — Pastel Flat - `glossy-glass-anime` — Glassy Anime - `grey-blocks-lineless` — Grey Wash - `hard-line-monochrome-accent` — Sharp Ink - `high-contrast-retro-vivid` — Neon Vivid - `holographic` — Hyper Pop - `korean-style` — Polished Manhwa - `lineart-watercolor-airy` — Luminous Wash - `low-saturation-expressive-eyes` — Muted & Transparent - `lucid-dreamy` — Lucid Dreamy - `luminous-retro-impasto` — Luminous Impasto - `minimal-flat-color` — Minimal Flat - `modern-oil-glamour` — Modern Glam Anime - `moe-pastel` — Moe Pastel - `oil-watercolor-blend` — Water-Infused Oil - `painterly-doodle` — Tender Brushstroke - `paper-watercolor` — Soft Aqueous - `pencil-sketch` — Pencil Sketch - `poster-graffiti` — Neon Poster - `retro-comic-fusion` — Retro Fusion - `retro-oil-glamour` — Retro Glam Anime - `retro-tv-anime` — 90s Retro-Cell - `semi-realistic` — Semi-Real - `stable-vaporwave` — Neon Vaporwave - `tinted-sketch` — Tinted Sketch - `vintage-collage` — Vintage Collage - type: object additionalProperties: false required: - type - custom properties: type: type: string enum: - custom custom: type: string description: | A raw style prompt string injected as a prefix to the generation prompt. This is an advanced option — incorrect values may degrade output quality. discriminator: propertyName: type loras: type: array maxItems: 15 description: | Apply one or more LoRA adapters to fine-tune output style or content. How many can be combined depends on your plan (3 without a membership, up to 15); more returns `403 LORA_NUM_EXCEEDED`. Use a LoRA made for the same base model as the generation model: a LoRA's page on PixAI shows its base model. How to find LoRA modelId: 1. Open the LoRA model page on PixAI. 2. Click the LoRA version you'd like to use. 3. Copy the last path segment of the URL — that's the LoRA version ID. Example URL: `https://pixai.art/model//` → LoRA version ID is ``. items: $ref: "#/components/schemas/LoraItem" contextImages: type: array maxItems: 3 x-badge: - type: tsubaki3 info: | Only available for **Tsubaki.3** (`modelVersionId` `2024383379556065549`). Other models return `400 BAD_REQUEST` ("This feature is not yet supported for this model"). description: | Reference images for in-context generation, e.g. keeping a character consistent or taking a pose from another image. Each item is a media ID: an ID from a finished task's `outputs.mediaIds`, or one returned by [Upload an image or video](/en/docs/api/media/uploadMedia). Each image must be under 4000 px on both sides. items: type: string description: A media ID. example: [ "1943547603237249965" ] seed: type: integer description: | Random seed for reproducible generation. Omit to use a randomly generated seed. example: 42 batchSize: type: integer enum: - 1 - 4 default: 1 description: | Number of images to generate in one task. Supported values are `1` and `4`. Defaults to `1`. example: 4 sampling: type: object x-badge: - type: sdxl-sdv1 info: | Only available for **SDXL** and **SD v1** model families. description: | Advanced sampling parameters for fine-tuning the diffusion process. All fields are optional — omitting a field uses the model's default. Refer to the [Models](/en/docs/references/models) page for allowed sampling methods per model. additionalProperties: false properties: method: type: string description: | Sampler algorithm used during generation. Supported methods vary by model — see the [Models](/en/docs/references/models) page. example: "DPM++ 2M Karras" steps: type: integer minimum: 1 maximum: 100 description: | Number of diffusion steps. Higher values can improve detail but increase generation time and cost. example: 25 cfgScale: type: number description: | Classifier-Free Guidance (CFG) scale. Controls the balance between prompt adherence and creative freedom. Higher values follow the prompt more strictly. example: 7 promptHelper: type: string enum: - enable - disable - low - medium default: enable description: | Whether to use the built-in prompt helper to enhance the prompt before generation. - `enable`: Automatically enhance the prompt (default). - `disable`: Use the prompt as-is. - `low` / `medium`: How freely the helper rewrites the prompt, on Tsubaki.2 and Tsubaki.3. Other models treat both as `enable`. x-callout: "Enabling this field will incur additional charges." callbackUrl: type: string format: uri description: | A publicly accessible URL to receive task status updates via POST request (webhook). We will send a POST request to the callback URL when the task status changes. example: "https://your-server.example.com/webhook" colorPalette: x-badge: - type: tsubaki3 info: | Only available for **Tsubaki.3** (`modelVersionId` `2024383379556065549`). Other models return `400 BAD_REQUEST` ("This feature is not yet supported for this model"). description: | Steer the colors of the image. Take a preset from `GET /v2/color-palettes/presets` and send it whole, with `"source": "preset"` added. allOf: - $ref: "#/components/schemas/ColorPalette" - type: object required: - source properties: source: type: string enum: - preset - user LoraItem: type: object required: - modelId additionalProperties: false properties: modelId: type: string description: The LoRA model version ID from the PixAI Model Market. example: "1744880666293972790" weight: type: number minimum: 0 maximum: 1 default: 1 description: The influence weight of the LoRA adapter. Range 0.0–1.0. example: 0.7 triggerWords: type: string description: | Custom trigger words for this LoRA. If omitted, the LoRA's default trigger words are added automatically; an empty string adds none. Error: type: object required: - code - status - message properties: code: type: string description: Machine-readable error code, e.g. `BAD_REQUEST` or `INSUFFICIENT_BALANCE`. Branch on this, not on `message`. status: type: integer description: The HTTP status code. message: type: string description: Human-readable detail, often naming the field or value at fault. data: type: object description: Code-specific detail, e.g. `issues` for validation errors or `reason` for `IMAGE_SIZE_NOT_ALLOWED`. defined: type: boolean description: Whether `code` is one this endpoint declares. Task: type: object required: - id - status - createdAt - updatedAt properties: id: type: string description: "The unique identifier for the task." status: type: string description: "The current status of the task." enum: - waiting - running - completed - cancelled - failed createdAt: type: string description: "The time the task was created." format: date-time updatedAt: type: string description: "The time the task was last updated." format: date-time startedAt: type: string description: "The time the task started running." format: date-time endAt: type: string description: "The time the task finished." format: date-time taskKind: type: string description: | What the task makes: `t2i` or `i2i` for images, `inpaint` for `POST /v2/gen-task/inpaint`, `i2v-inhouse` for `POST /v2/frame-video/create`, `reference-video` for `POST /v2/reference-video/create`. paidQuota: type: integer description: Credits charged for the task. taskStatusMsgI18n: type: object description: Why a `failed` task failed. properties: key: type: string description: A stable message key. defaultMessage: type: string description: The message in English. outputs: type: object description: "The outputs produced by the task." properties: videoMediaId: type: string description: | For video tasks, the media ID of the video. Resolve it with `GET /v2/media/batch?ids=` and download `fileUrl`. `mediaIds` then holds only the poster image. seed: type: string description: The seed used, as a decimal string. mediaIds: type: array description: | Media IDs of the generated images. Use these IDs to fetch detailed information about each generated image. Generated images are not permanently retained — retrieve them as soon as possible. items: type: string description: The unique identifier for a generated media item. mediaUrls: type: array description: | Public URLs of the generated images. Generated images are not permanently retained — retrieve them as soon as possible. If an image is unavailable, its entry may be `null`. items: type: - string - "null" description: The URL of the media item, or null if unavailable. UploadUrlRequest: type: object properties: isEphemeral: type: boolean default: false description: Mark the file as a short-lived generation input. Ephemeral files are private, and their URLs are signed and expire. UploadUrl: type: object required: - uploadUrl - externalId properties: uploadUrl: type: string description: Presigned URL. `PUT` the raw file bytes to it with the file's `Content-Type`. externalId: type: string description: Pass this to the matching `register` endpoint after the upload. Media: type: object required: - id - type properties: id: type: string description: The media ID. type: type: string enum: - IMAGE - VIDEO - AUDIO urls: type: array description: Image URLs by variant. `PUBLIC` is the full image. items: type: object properties: variant: type: string enum: - PUBLIC - THUMBNAIL - STILL_THUMBNAIL url: type: string width: type: - number - "null" height: type: - number - "null" fileUrl: type: - string - "null" description: URL of the file itself, e.g. the MP4 of a video. hlsUrl: type: - string - "null" description: HLS stream of a video. Often `null` for generated videos; use `fileUrl`. thumbnailUrl: type: - string - "null" description: Poster image of a video. Often `null` for generated videos; the task's `outputs.mediaIds` holds the poster. duration: type: - number - "null" description: Length of a video or audio file in seconds. Can be `0` for generated videos. ModelVersion: type: object required: - id - modelId properties: id: type: string description: The version ID. Pass it as `modelVersionId`, or as `loras[].modelId` for a LoRA. modelId: type: string description: The ID of the model this version belongs to. modelTitle: type: string name: type: string description: Version name. modelType: type: - string - "null" description: e.g. `SDXL_MODEL`, `MMDIT26B_MODEL` or `SDXL_LORA`. loraBaseModelType: type: - string - "null" description: For a LoRA, the base model type it was trained for. Use a base model of this type. loraBaseModelId: type: - string - "null" description: For a LoRA, the base model it was trained on. status: type: - string - "null" description: Only `available` versions can be used for generation. extra: type: object description: Version details. `triggerWords` holds the default trigger words of a LoRA. properties: triggerWords: type: string mediaId: type: - string - "null" description: Cover image media ID. ModelListItem: type: object required: - id - title - type properties: id: type: string description: The model ID. Generation endpoints take a version ID instead; use `latestAvailableVersionId`. title: type: string type: type: string description: e.g. `SDXL_MODEL`, `MMDIT26B_MODEL` or `SDXL_LORA`. category: type: - string - "null" loraBaseModelTypes: type: array items: type: string description: For a LoRA, the base model types it works with. permittedUse: type: string enum: - PERSONAL - COMMERCIAL description: What the license allows. mediaId: type: - string - "null" description: Cover image media ID. latestAvailableVersionId: type: - string - "null" description: The newest usable version. Pass it as `modelVersionId`, or as `loras[].modelId` for a LoRA. latestAvailableVersion: type: - object - "null" description: The newest usable version. properties: id: type: string description: Same as `latestAvailableVersionId`. name: type: string modelType: type: - string - "null" loraBaseModelType: type: - string - "null" description: For a LoRA, the base model type it was trained for. extra: type: object description: "`triggerWords` holds the default trigger words of a LoRA." properties: triggerWords: type: string ColorPalette: type: object required: - id - name - palette - createdAt - updatedAt properties: id: type: string name: type: string palette: type: object description: Colors for the whole image and, optionally, the background and characters. `overall` or `background` is required. properties: overall: $ref: "#/components/schemas/PaletteColors" background: $ref: "#/components/schemas/PaletteColors" character: $ref: "#/components/schemas/PaletteColors" createdAt: type: string format: date-time updatedAt: type: string format: date-time PaletteColors: type: object required: - colors properties: colors: type: array minItems: 1 maxItems: 12 items: type: object required: - hex - ratio properties: hex: type: string description: "`#RRGGBB`." ratio: type: number minimum: 0 maximum: 100 description: Share of the image, in percent. securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: API-Key description: | In all requests that require authentication headers: X-Trace-Id: description: "A unique identifier for the request. Useful for debugging and tracing." schema: type: string responses: Unauthorized: description: '`UNAUTHORIZED`: no API key was sent. An invalid, revoked or expired key answers `{"message": "invalid token", "code": 40100001}` instead, with a numeric `code`.' content: application/json: schema: $ref: "#/components/schemas/Error" CannotRun: description: | The account cannot run this task right now. `code` says why: - `INSUFFICIENT_BALANCE`: not enough credits for the task. - `WAITING_NUM_EXCEEDED`: too many of your tasks are still `waiting` (`data.currentWaitingTasks`). Wait for some to start before submitting more. - `REQUIRE_MEMBERSHIP`: the feature needs a PixAI membership (image-to-video). - `NSFW_PROMPTS_DETECTED`: the prompt was refused because the account is in safe mode. content: application/json: schema: $ref: "#/components/schemas/Error" CreationFailed: description: "`CREATION_FAILED`: the task could not be created. Retry with exponential backoff." content: application/json: schema: $ref: "#/components/schemas/Error" ModelNotFound: description: "`NOT_FOUND`: no such model or version, or it is private. The two cases are not told apart." content: application/json: schema: $ref: "#/components/schemas/Error"