> ## Documentation Index > Fetch the complete documentation index at: https://docs.nano-gpt.com/llms.txt > Use this file to discover all available pages before exploring further. # Video Generation > Generate videos using supported text-to-video, image-to-video, and video-to-video models. The response includes a runId and pending status; poll the status endpoint for completion. See the docs for the current model list and required inputs. > Image-conditioned models accept either `imageDataUrl` (base64) or a public `imageUrl`. The service uses the explicit value you provide before checking any saved attachments. ## Overview `POST /generate-video` submits an asynchronous job to create, extend, or edit a video. The endpoint responds immediately with `runId`, `id`, `model`, and `status: "pending"`. `runId` and `id` are the same NanoGPT job identifier (format `vid_...`). Poll the unified Video Status endpoint with that job ID until you receive final assets. Duration-based billing is assessed after completion. Errors include descriptive JSON payloads. Surface the `error.message` (and HTTP status) to help users correct content-policy or validation issues. ## Extend Workflows * **Midjourney extend (task-based)**: use `POST /api/generate-video/extend` with `runId` (preferred) or `taskId` (legacy alias) plus `index` (0-3). This flow does not accept `video`, `videoUrl`, `videoDataUrl`, or `videoAttachmentId`. * **Source video extend (extend models)**: use `POST /api/generate-video` with an extend model plus `prompt` and a source video (`videoUrl`, `videoDataUrl`, or `videoAttachmentId`). `video` is only accepted by select models (for example, `wan-wavespeed-25-extend`). Max source video length: 120 seconds. ## Request Schema Only include the fields required by your chosen `model`. Unknown keys are ignored, but some models fail when extra media fields are present. ### Core Fields | field | type | required | details | | - | - | - | - | | `model` | string | yes | Video model ID. Model availability changes; discover models via `GET /api/v1/models?detailed=true`. | | `conversationUUID` | string | no | Attach the request to a conversation thread. | | `prompt` | string | conditional | Required for text-to-video and edit models unless a structured script is supplied. | | `negative_prompt` | string | no | Suppresses specific content. Respected by Veo, Wan, Runway, Pixverse, and other models noted below. | | `script` | string | conditional | LongStories models accept full scripts instead of relying on `prompt`. | | `storyConfig` | object | conditional | LongStories structured payload (e.g. scenes, narration, voice). | | `animation` | boolean | no | Enables animation for LongStories outputs. | | `language` | string | no | Output language for LongStories. | | `characters` | array | no | Character definitions for LongStories. | | `duration` | string | conditional | Seconds as a string (`"5"`, `"8"`, `"60"`). Limits vary per model; see individual entries. | | `seconds` | string | conditional | Sora-specific duration selector (`"4"`, `"8"`, `"12"`). | | `aspect_ratio` | string | conditional | Ratios such as `16:9`, `9:16`, `1:1`, `3:4`, `4:3`, `21:9`, `auto`. | | `orientation` | string | conditional | `landscape` or `portrait` for Sora and Wan text/image flows. | | `resolution` | string | conditional | Resolution tokens (`480p`, `580p`, `720p`, `1080p`, `1792x1024`, `2k`, `4k`). | | `size` | string | conditional | Output size preset (supported by select models). | | `mode` | string | no | Operation mode: `text-to-video`, `image-to-video`, `reference-to-video`, `video-edit`. | | `generateAudio` | boolean | no | Adds AI audio on Veo 3 and Lightricks models. Defaults to `false`. | | `enhancePrompt` | boolean | no | Optional Veo 3 prompt optimizer. Defaults to `false`. | | `pro_mode` / `pro` | boolean | no | High-quality toggle for Sora and Hunyuan families. Defaults to `false`. | | `enable_prompt_expansion` | boolean | no | Prompt booster for Wan/Seedance/Minimax variants. Disabled by default. | | `enable_safety_checker` | boolean | no | Optional safety checker toggle (supported by select models). | | `camera_fix` / `camera_fixed` / `cameraFixed` | boolean | no | Locks the virtual camera for Seedance and Wan variants. | | `seed` | number or string | no | Optional seed forwarded on model/provider routes that support it. This may improve reproducibility but does not guarantee identical output. | | `voiceId` | string | conditional | Alternate voice selector for lipsync models. | | `voice_id` | string | conditional | Required by `kling-lipsync-t2v`. | | `voice_language` | string | conditional | `en` or `zh` for `kling-lipsync-t2v`. | | `voice_speed` | number | conditional | Range `0.8-2.0` for `kling-lipsync-t2v`. | | `videoDuration` / `billedDuration` | number | no | Optional overrides for upscaler billing calculations. | | `adjust_fps_for_interpolation` | boolean | no | Optional toggle for interpolation-aware upscaling. Defaults to `false`. | ### Media Inputs | field | type | required | details | | - | - | - | - | | `imageDataUrl` | string | conditional | Base64-encoded data URL. Recommended for private assets or files larger than 4 MB. | | `imageUrl` | string | conditional | HTTPS link to a source image. | | `imageAttachmentId` | string | conditional | Reference to a library-stored image. | | `image` | string | conditional | Alternate image field accepted by select models. Prefer `imageUrl` unless the model explicitly requires `image`. | | `reference_image` | string | conditional | Optional still image guiding `runwayml-gen4-aleph`. | | `referenceImages` | array | conditional | Multiple reference images for reference-to-video flows. | | `referenceVideos` | array | conditional | Multiple reference videos. | | `audioDataUrl` | string | conditional | Base64 data URL for audio-driven models. | | `audioDuration` | number | conditional | Duration of provided audio in seconds. | | `audioUrl` | string | conditional | HTTPS audio input. | | `audio` | string | conditional | Alternate audio field accepted by select models. Prefer `audioUrl` unless the model explicitly requires `audio`. | | `videoUrl` | string | conditional | HTTPS link to a source video (edit, extend, upscaler, or lipsync jobs). | | `videoDataUrl` | string | conditional | Base64 data URL for a source video. | | `video` | string | conditional | Alternate video field accepted by select models. Prefer `videoUrl` unless the model explicitly requires `video`. | | `videoAttachmentId` | string | conditional | Reference to a library-stored video. | | `swapImage` | string | conditional | Swap image (face-swap models). | | `targetVideo` | string | conditional | Target video (face-swap models). | | `targetFaceIndex` | number | no | Optional face index (face-swap models). | > Provide only the media fields that your target model expects. Extra media inputs often trigger validation errors. > Prefer `videoUrl` (camelCase) for source videos; only send `video` when the model explicitly requires it. ### Advanced Controls | field | type | models | | - | - | - | | `num_frames` | integer | Wan 2.2 families, Seedance 22 5B, Wan image-to-video. | | `frames_per_second` | integer | Wan 2.2 5B. | | `num_inference_steps` | integer | Wan 2.2 families. | | `guidance_scale` | number | Wan 2.2 5B. | | `shift` | number | Wan 2.2 5B. | | `interpolator_model` | string | Wan 2.2 5B. | | `num_interpolated_frames` | integer | Wan 2.2 5B. | | `movementAmplitude` | string | Select models (for example `auto`, `small`, `medium`, `large`). | | `motion` | string | Select models (for example `low`, `high`). | | `style` | string | Select models (style/preset strings). | | `effectType`, `effect`, `cameraMovement`, `motionMode`, `soundEffectSwitch`, `soundEffectPrompt` | varies | Pixverse v4.5/v5. | | `mode` | string | Select models (for example `animate`, `replace`). | | `prompt_optimizer` | boolean | Select models. | ## Model Discovery Video model IDs and supported fields change over time. Use `GET /api/v1/models?detailed=true` to discover the current list and select a model intended for video generation. Notes: * Different models accept different media inputs (for example `imageUrl` vs a source `videoUrl`) and may support different duration / resolution options. * If you see validation errors, first retry with only the minimal required fields for your chosen model. ## Async Processing & Status Polling * The submission response includes `{ runId, id, model, status: "pending" }` where `id` and `runId` are identical. * Poll `/api/video/status?requestId=` (or `runId`) until the job reaches `status: "COMPLETED"` or `status: "FAILED"`. The legacy `/api/generate-video/status` endpoint is deprecated. * Many jobs emit intermediate states (`queued`, `processing`, `generating`, `delivering`). Persist them if you need audit trails. * Failed jobs include an `error` object. Surface the message and adjust prompts or inputs before retrying. * Duration and resolution determine credit usage. ### Response example ```json theme={null} { "runId": "vid_m1abc123def456", "id": "vid_m1abc123def456", "status": "pending", "model": "veo2-video", "cost": 0.35, "paymentSource": "XNO", "remainingBalance": 12.5, "prechargeLabel": "string" } ``` ## Content & Safety Notes Some models may block prompts that violate content policies. Non-200 responses describe the violation reason; relay these messages verbatim to users or implement automated prompt adjustments. ## Next Steps * Poll the Video Status endpoint after every submission to retrieve final assets. * Keep customer-facing pricing tables in sync with the API behavior you observe in production. ## OpenAPI ````yaml POST /generate-video openapi: 3.1.0 info: title: NanoGPT API description: >- API documentation for the NanoGPT language, image, video, speech-to-text, and text-to-speech generation services license: name: MIT version: 1.0.0 servers: - url: https://api.nano-gpt.com/api description: >- Direct API for uploads and long-running requests; choose a model listed by this host - url: https://nano-gpt.com/api description: Website API for current text model examples security: [] paths: /generate-video: post: description: >- Generate videos using supported text-to-video, image-to-video, and video-to-video models. The response includes a runId and pending status; poll the status endpoint for completion. See the docs for the current model list and required inputs. requestBody: description: Parameters for video generation across different models and providers content: application/json: schema: $ref: '#/components/schemas/VideoGenerationRequest' required: true responses: '202': description: >- Video generation request submitted successfully (asynchronous processing) content: application/json: schema: $ref: '#/components/schemas/VideoGenerationResponse' '400': description: Bad Request - Invalid parameters or safety filter triggered content: application/json: schema: $ref: '#/components/schemas/Error' '401': description: Unauthorized - Invalid or missing API key '500': description: Server Error - Video generation failed content: application/json: schema: $ref: '#/components/schemas/Error' security: - apiKeyAuth: [] components: schemas: VideoGenerationRequest: type: object required: - model properties: model: type: string description: >- The video model to use for generation. See the docs for the current model list and required inputs. prompt: type: string description: Text prompt describing the video to generate example: A serene lake at sunset with gentle ripples on the water script: type: string description: >- Fully-written script for LongStories models (takes precedence over prompt) conversationUUID: type: string description: UUID for conversation tracking projectId: type: string description: Project identifier for LongStories models framework: type: string description: Story framework for LongStories models enum: - default - emotional_story - product_showcase - tutorial default: default shortRequestEnhancer: type: boolean description: >- Smart Enhancement: if true, automatically choose better framework and add Director Notes if necessary default: false targetLengthInWords: type: integer description: Target length in words for LongStories models (legacy parameter) default: 70 targetLengthInSeconds: type: integer description: Target length in seconds (alternative to words) directorNotes: type: string description: >- Prompt for the image generation engine (LongStories). Example: 'Warm lighting' or 'Make the first image very impactful' example: Warm, cozy lighting with focus on people interacting aspectRatio: type: string description: Video aspect ratio for LongStories enum: - '9:16' - '16:9' default: '9:16' scriptConfig: type: object description: Script generation configuration for LongStories properties: style: type: string description: Sets the tone & voice of the generated script enum: - default - no_style - engaging_conversational - dixit_biography - kind_biography - hero_journey - emotional_story - dramatic_reveal - heartwarming_stories - educational_history - news_brief default: default targetLengthInSeconds: type: integer description: >- Approx. time length of the script in seconds. The actual length may vary slightly. minimum: 5 maximum: 600 default: 30 imageConfig: type: object description: Image generation configuration for LongStories properties: model: type: string description: Which image model to use for generating images enum: - flux_schnell - flux_lora - flux_pro - flux_pro_ultra - flux-kontext - recraft - sdxl - sdxl_lora - sd35_large - sd35_medium - sd35_large_turbo - leonardo_flux_precision - leonardo_phoenix_quality - leonardo_phoenix_ultra - reve-v1 - hidream_fast - hidream_dev - hidream_full - gpt_image_1 - imagen4_preview - midjourney default: hidream_dev loraConfig: type: object description: Style configuration (only applies when using Flux LoRA model) properties: loraSlug: type: string description: Specialized style to apply enum: - ghibsky-comic-book - colour-sketches - sketch-paint - 90s-anime - 2000s-crime-thrillers - xno-symbol-flux default: ghibsky-comic-book videoConfig: type: object description: Video generation configuration for LongStories properties: enabled: type: boolean description: Generate video motion from images instead of using static images default: true model: type: string description: Which video model to use for motion generation enum: - ltx_video_13b_distilled - ltx_video - stable_video - motion - kling_v1_6_std_5s - kling_v2_1_std_5s - ltx_video_13b_dev - skyreels - vidu - kling_v1_6_std_10s - kling_v2_1_std_10s - kling_v2_1_pro_5s - luma_ray2 - kling_pro_5s - minimax - wan_i2v - motion_2 - kling_v2_1_pro_10s - wan_pro - luma_ray2_hd - kling_pro_10s - kling_v2_1_master_5s - kling_v2_master_5s - veo2 - kling_v2_1_master_10s - kling_v2_master_10s default: kling_v2_1_std_5s voiceoverConfig: type: object description: Voiceover configuration for LongStories properties: enabled: type: boolean description: Enable AI voiceover for the video default: true voiceId: type: string description: Voice for video narration enum: - 9BWtsMINqrJLrRacOk9x - CwhRBWXzGAHq8TQ4Fs17 - EXAVITQu4vr4xnSDxMaL - FGY2WhTYpPnrIDTdsKH5 - IKne3meq5aSn9XLyUdCD - JBFqnCBsd6RMkjVDRZzb - N2lVS1w4EtoT3dr4eOWO - SAz9YHcvj6GT2YYXdXww - TX3LPaxmHKxFdv7VOQHJ - XB0fDUnXU5powFXDhCwa - Xb7hH8MSUJpSbSDYk0k2 - XrExE9yKIg1WjnnlVkGX - bIHbv24MWmeRgasZH58o - cgSgspJ2msm6clMCkdW9 - cjVigY5qzO86Huf0OWal - nPczCjzI2devNBz1zQrb - onwK4e9ZLuTAKqWW03F9 - pqHfZKP75CvOlQylNhV4 - pFZP5JQG7iQjIQuC4Bku - KHCvMklQZZo0O30ERnVn - Nh2zY9kknu6z4pZy6FhD - LlZr3QuzbW4WrPjgATHG - YExhVa4bZONzeingloMX - m1VE7dnwBN0zMer3LcKv - zWDA589rUKXuLnPRDtAG - YYHkBdgrAwQWIaH6m2ai default: zWDA589rUKXuLnPRDtAG captionsConfig: type: object description: Captions configuration for LongStories properties: captionsEnabled: type: boolean description: Show text captions in the video default: true captionsStyle: type: string description: Style of video captions enum: - default - minimal - neon - cinematic - fancy - tiktok - highlight - gradient - intellectual - vida - manuscripts - subtitle - modern - bounce - popcorn - typewriter - handwritten - karaoke - retro - gaming default: tiktok effectsConfig: type: object description: Effects configuration for LongStories properties: transition: type: string description: Transition style between different images enum: - fade - random - slide - wipe - flip - none default: fade floating: type: boolean description: Make the images move around slightly with floating effects default: true musicConfig: type: object description: Music configuration for LongStories properties: enabled: type: boolean description: Add background music to the video default: true musicSlug: type: string description: Choose background music for your video enum: - '' - temple_of_treasures - gentle_ambient_loop - serene_ambience - soothing_ambience - soothing_ambient_backdrop - tranquil_ambience - dreamscape - belonging_resonance - vivid_memories - cinematic_intro - cinematic_teaser - dramatic_cinematic_score - thriller_cinema_trailer - fractured_paintings - promise_of_tomorrow - spooky_orchestral_theme - light_upbeat_melody - puzzle_time - stomping_drums_rhythm - stomps_and_claps_rhythm_track - news_theme - adventurous_intro - burlesque_sweetheart - highway_nocturne_national_sweetheart - haptic_sensation default: gentle_ambient_loop volume: type: number description: Volume level for background music minimum: 0 maximum: 1 default: 0.3 loop: type: boolean description: Whether to loop the background music default: true voice: type: string description: 'Legacy: Voice ID for narration (use voiceoverConfig.voiceId instead)' example: pNInz6obpgDQGcFmaJgB captionsShow: type: boolean description: >- Legacy: Whether to show captions (use captionsConfig.captionsEnabled instead) default: true captionsStyle: type: string description: >- Legacy: Style for captions (use captionsConfig.captionsStyle instead) enum: - default - minimal - neon - cinematic - fancy - tiktok - highlight - gradient - instagram - vida - manuscripts default: default effects: type: object description: 'Legacy: Video effects configuration (use effectsConfig instead)' properties: transition: type: string default: fade floating: type: boolean default: false quality: type: string description: 'Legacy: Video quality (handled by videoConfig now)' enum: - low - medium - high default: medium motion: type: object description: 'Legacy: Motion configuration (handled by videoConfig now)' properties: enabled: type: boolean default: false strength: type: integer minimum: 1 maximum: 10 default: 3 music: type: string description: 'Legacy: Music track (use musicConfig instead)' example: video-creation/music/dramatic_cinematic_score.mp3 duration: oneOf: - type: string - type: integer description: >- Video duration (format varies by model - '5s' for Veo2, '5' for Kling, etc.) example: 5s aspect_ratio: type: string description: Aspect ratio (supported by select models) enum: - '16:9' - '9:16' - '1:1' - '4:3' - '3:4' default: '16:9' negative_prompt: type: string description: Negative prompt to avoid certain elements example: blur, distort, and low quality cfg_scale: type: number description: Classifier-free guidance scale minimum: 0 maximum: 20 default: 0.5 imageDataUrl: type: string description: >- Base64 data URL of input image for image-to-video models. Aliases `image_data_url` and `image` are also accepted and normalized. example: data:image/jpeg;base64,/9j/4AAQ... imageUrl: type: string description: >- Public HTTPS URL of the input image (interchangeable with imageDataUrl). The service will prioritize whichever field you supply before falling back to library attachments. example: https://images.unsplash.com/photo-1504196606672-aef5c9cefc92?w=1024 imageAttachmentId: type: string description: Library attachment ID for input image videoUrl: type: string description: >- Public HTTPS URL of the input video (extend/edit/upscale). Preferred field name for source videos. videoDataUrl: type: string description: Base64 data URL of the input video. video: type: string description: Alternate video field accepted by select providers. videoAttachmentId: type: string description: Library attachment ID for input video. prompt_optimizer: type: boolean description: Whether to optimize the prompt (MiniMax model) default: true num_inference_steps: type: integer description: Number of inference steps minimum: 1 maximum: 50 default: 30 pro_mode: type: boolean description: Enable pro mode for Hunyuan Video default: false resolution: type: string description: Video resolution enum: - 720p - 1080p - 540p default: 720p num_frames: oneOf: - type: integer - type: string description: Number of frames to generate default: 81 frames_per_second: type: integer description: Frames per second minimum: 5 maximum: 24 default: 16 seed: type: integer description: >- Optional seed forwarded on model/provider routes that support it. This may improve reproducibility but does not guarantee identical output. enable_safety_checker: type: boolean description: Enable safety content filtering default: true showExplicitContent: type: boolean description: Allow explicit content (inverse of safety checker) default: false enable_prompt_expansion: type: boolean description: Enable automatic prompt expansion acceleration: type: boolean description: Enable acceleration for faster processing shift: type: number description: Shift parameter for certain models age_slider: type: integer description: Age setting for PromptChan model minimum: 18 maximum: 60 default: 18 audioEnabled: type: boolean description: Enable audio for PromptChan model default: false video_quality: type: string description: Video quality for PromptChan model enum: - Standard - High default: Standard aspect: type: string description: Aspect setting for PromptChan model enum: - Portrait - Landscape - Square default: Portrait VideoGenerationResponse: type: object required: - runId - status - model properties: runId: type: string description: Unique identifier for the video generation request projectId: type: string description: Project identifier (for LongStories models) status: type: string description: Current status of the generation enum: - pending - processing - completed - failed default: pending model: type: string description: The model used for generation cost: type: number description: Cost of the video generation paymentSource: type: string description: Payment source used (USD or XNO) remainingBalance: type: number description: Remaining balance after the generation prechargeLabel: type: string description: Provider label for the precharge Error: required: - error - message type: object properties: error: type: integer format: int32 message: type: string securitySchemes: apiKeyAuth: type: apiKey in: header name: x-api-key ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.