> ## Documentation Index > Fetch the complete documentation index at: https://docs.longstories.ai/llms.txt > Use this file to discover all available pages before exploring further. # Generate a Video > Initiates an AI-assisted video generation (up to 10 minutes) using either a text prompt (for script + visuals) or a user-provided script (to skip script generation). Returns a Run ID used to query the status or results of the generation process. ## Base URL ```bash theme={null} https://longstories.ai/api/v1 ``` ## Check Styles, Music, Captions... You can check the styles, music, captions, effects, and other parameters you have available for your video generation at [longstories.ai/resources](https://longstories.ai/resources). **Note:** Despite the endpoint being named `/short`, this API supports generating videos up to 10 minutes in length. ## OpenAPI ````yaml POST /short openapi: 3.0.3 info: title: Video Generation API version: 1.0.0 description: >- Generate AI-powered videos (up to 10 minutes) from text prompts or existing scripts. servers: - url: https://longstories.ai/api/v1 description: Production API Server security: [] paths: /short: post: summary: Create a Video description: >- Initiates an AI-assisted video generation (up to 10 minutes) using either a text prompt (for script + visuals) or a user-provided script (to skip script generation). Returns a Run ID used to query the status or results of the generation process. operationId: createShortVideo requestBody: description: >- Parameters used to initiate video creation. Provide at least one of `prompt` or `script`. required: true content: application/json: schema: $ref: '#/components/schemas/ShortGenerationRequest' example: prompt: >- Tell a heartwarming story about a small bakery that becomes the heart of its community shortRequestEnhancer: false directorNotes: >- Warm, cozy lighting with focus on people interacting in the bakery scriptConfig: style: default targetLengthInWords: 70 imageConfig: model: flux_schnell voiceoverConfig: enabled: true voiceId: zWDA589rUKXuLnPRDtAG captionsConfig: captionsEnabled: true captionsStyle: tiktok effectsConfig: transition: fade floating: true musicConfig: musicSlug: gentle_ambient_loop enabled: true volume: 0.3 loop: true videoConfig: enabled: false model: luma_ray2 templateConfig: templateId: none quality: medium aspectRatio: '9:16' responses: '200': description: >- Creation request accepted. Returns a generation Run ID used to fetch status or results later. content: application/json: schema: allOf: - $ref: '#/components/schemas/SuccessResponse' - properties: data: $ref: '#/components/schemas/GenerationRunId' example: data: id: run_abc123 projectId: proj_123abc requestId: req_123abc '400': description: >- Invalid input or missing required fields. This could happen if neither `prompt` nor `script` was provided, or if a field type is invalid (e.g., `quality: "ultra"`). content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: MissingPrompt: summary: Missing prompt or script value: error: code: VAL_MISSING_FIELD message: Either prompt or script is required details: {} InvalidValue: summary: Unrecognized quality parameter value: error: code: VAL_INVALID_FORMAT message: Invalid format provided details: field: quality reason: 'Must be one of: high, medium, low' '401': description: 'Unauthorized: invalid or missing API key in `x-api-key` header.' content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' example: error: code: AUTH_INVALID_KEY message: Invalid API key provided requestId: req_789ghi '402': description: Insufficient credits to generate the video. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' example: error: code: BIZ_INSUFFICIENT_CREDITS message: Insufficient credits to perform this operation requestId: req_012jkl details: required: null available: 25 '429': description: Rate limit exceeded (too many requests). content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' example: error: code: RATE_TOO_MANY_REQUESTS message: Too many requests, please try again later requestId: req_345mno details: retryAfter: 60 '500': description: >- Internal server error or unexpected error encountered during generation. Could be a rendering issue, hardware problem, or an uncaught exception. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' example: error: code: BIZ_GENERATION_FAILED message: Error during video generation requestId: req_456def details: reason: Resource allocation failed security: - ApiKeyAuth: [] components: schemas: ShortGenerationRequest: type: object oneOf: - required: - prompt - required: - script properties: prompt: type: string description: >- A textual description of what you want your video to depict or explain. Used to generate the script, scenes, and voiceover. MUTUALLY EXCLUSIVE with 'script' field - only provide one of 'prompt' or 'script'. script: type: string description: >- A fully-written script to skip AI script generation. MUTUALLY EXCLUSIVE with 'prompt' field - only provide one of 'prompt' or 'script'. scriptConfig: $ref: '#/components/schemas/ScriptConfig' directorNotes: type: string description: >- Prompt for the image generation engine. Example: 'Warm lighting' or 'Make the first image very impactful' shortRequestEnhancer: type: boolean description: >- Experimental: if true, we choose a better framework for you and add Director Notes if necessary. It doesn't overwrite your settings. default: false imageConfig: $ref: '#/components/schemas/ImageConfig' voiceoverConfig: $ref: '#/components/schemas/VoiceoverConfig' captionsConfig: $ref: '#/components/schemas/CaptionsConfig' effectsConfig: $ref: '#/components/schemas/EffectsConfig' musicConfig: $ref: '#/components/schemas/MusicConfig' videoConfig: $ref: '#/components/schemas/VideoConfig' templateConfig: type: object properties: templateId: type: string enum: - none - longstories default: none description: >- Internal use only - Template identifier for video outro and other pre-sets. description: Internal use only - Template configuration for LongStories branding. quality: type: string enum: - high - medium - low default: medium description: >- Video resolution & encoding quality. Higher quality takes longer to render and results in a larger file size. Note: Different quality levels will have different costs in the future, but currently only affect file size and visual quality. aspectRatio: type: string enum: - '9:16' - '16:9' description: Video aspect ratio - vertical (9:16) or horizontal (16:9) default: '9:16' SuccessResponse: type: object description: Generic success envelope containing data and a request identifier. properties: data: type: object description: The primary payload returned by the endpoint (if applicable). requestId: type: string description: >- Unique identifier associated with the request for debugging/tracking. required: - data GenerationRunId: type: object description: >- Return value for a newly created short generation run. Use these values to check progress. properties: id: type: string description: >- Unique run identifier. Append as `?runId={id}` to `GET /short` to fetch status. projectId: type: string description: Unique identifier for the project associated with this generation. ErrorResponse: properties: error: type: object properties: code: type: string description: >- Short code describing the error type (e.g., VAL_INVALID_FORMAT, AUTH_INVALID_KEY). message: type: string description: Human-readable explanation of the error. details: type: object description: >- Additional diagnostic info, such as which field is invalid or how many credits are needed. requestId: type: string description: >- Unique identifier for debugging. Provided in logs for cross-reference. required: - code - message required: - error ScriptConfig: type: object description: >- Choose in which Style the script is written, and how long it should be. Only works if `script` is not provided. See [this page](https://longstories.ai/resources/script-styles) for more information. properties: style: type: string enum: - default - no_style - engaging_conversational - dixit_biography - kind_biography - hero_journey - emotional_story - dramatic_reveal - heartwarming_stories - educational_history - news_brief default: default description: >- Sets the tone & voice of the generated script. Examples: 'storytelling' for narrative or 'educational' for facts/instructional tone. targetLengthInWords: type: integer default: 70 minimum: 1 maximum: 1500 description: >- Approx. word length of the script. The actual length may vary slightly. ImageConfig: type: object description: >- Parameters for still-image generation used as video backgrounds or scenes. properties: model: type: string enum: - flux_schnell - imagen4_fast - flux_pro - flux_pro_ultra - recraft - flux_lora - sdxl - sdxl_lora - bytedance_seedream_v3 - sd35_large - sd35_medium - sd35_large_turbo - leonardo_flux_precision - leonardo_phoenix_quality - reve-v1 - leonardo_phoenix_ultra - hidream_fast - hidream_dev - hidream_full - midjourney - gpt_image_1 - imagen4_preview default: flux_schnell description: >- Which image generation model to use. Different models can produce different artistic or realistic styles. recraftStyle: type: string enum: - any - realistic_image - digital_illustration - realistic_image/b_and_w - realistic_image/hard_flash - realistic_image/hdr - realistic_image/natural_light - realistic_image/studio_portrait - realistic_image/enterprise - realistic_image/motion_blur - digital_illustration/pixel_art - digital_illustration/hand_drawn - digital_illustration/grain - digital_illustration/infantile_sketch - digital_illustration/2d_art_poster - digital_illustration/handmade_3d - digital_illustration/hand_drawn_outline - digital_illustration/engraving_color - digital_illustration/2d_art_poster_2 description: >- If using the 'recraft' model, you can choose a preset style. Ignored if `model` is not 'recraft'. loraConfig: type: object description: >- Will only be applied if `model` is 'flux_lora' or 'sdxl_lora'. Allows more specialized styles. properties: loraSlug: type: string enum: - ghibsky-comic-book - colour-sketches - sketch-paint - 90s-anime - 2000s-crime-thrillers - xno-symbol-flux description: >- Identifier or slug for the LoRA to apply. See possible slugs on [LoRA styles](https://longstories.ai/resources/lora-styles) for more information. leonardoStyle: type: string enum: - NONE - THREE_D_RENDER - ACRYLIC - ANIME_GENERAL - BOKEH - CINEMATIC - CINEMATIC_CONCEPT - CREATIVE - DYNAMIC - FASHION - GAME_CONCEPT - GRAPHIC_DESIGN_3D - GRAPHIC_DESIGN_POP_ART - GRAPHIC_DESIGN_VECTOR - HDR - ILLUSTRATION - MACRO - MINIMALIST - MOODY - PORTRAIT - PORTRAIT_CINEMATIC - PRO_BW_PHOTOGRAPHY - PRO_COLOR_PHOTOGRAPHY - PRO_FILM_PHOTOGRAPHY - PORTRAIT_FASHION - RAY_TRACED - SKETCH_BW - SKETCH_COLOR - STOCK_PHOTO - VIBRANT - WATERCOLOR default: NONE description: >- Optional style preset for Leonardo models (leonardo_flux_precision, leonardo_phoenix_quality, leonardo_phoenix_ultra). Ignored for other models. Provide the style name (e.g., "CINEMATIC"). Compatibility varies per Leonardo model (e.g., ACRYLIC is Flux-only, BOKEH is Phoenix-only). Check model compatibility before use. VoiceoverConfig: type: object description: Enables and customizes an AI voiceover for the generated script content. properties: enabled: type: boolean default: true description: >- If false, no voiceover is produced. The final video will be music-only (if `musicConfig.enabled`) or silent. voiceId: type: string default: zWDA589rUKXuLnPRDtAG description: >- Which voice model to use from the AI voice library. E.g., 'zWDA589rUKXuLnPRDtAG' is a US male voice. Find all the available voices on [this page](https://longstories.ai/resources/voices). CaptionsConfig: type: object description: >- Controls visual styling and position of text captions derived from the voiceover. properties: captionsEnabled: type: boolean default: true description: >- If true, on-screen text captions are displayed following the voiceover text. captionsStyle: type: string enum: - default - minimal - neon - cinematic - fancy - tiktok - highlight - gradient - intellectual - vida - manuscripts - subtitle - modern - bounce - popcorn - typewriter - handwritten - karaoke - retro - gaming default: default description: >- How the captions appear visually (font, color, styling). See [this page](https://longstories.ai/resources/captions-styles) to see examples. captionsPosition: type: string enum: - top - center - bottom default: bottom description: Where to place the captions in the video. EffectsConfig: type: object description: >- Controls visual transitions and animations applied between scenes. 'Random' picks a dynamic effect per scene. properties: transition: type: string enum: - none - fade - slide - wipe - flip - zoom-in - zoom-out - sway-in - SpinClockwise - SpinCounterClockwise - SpinBoomerang - PendulumLeft - PendulumRight - TrainLeft - TrainRight - FallLeft - FallRight - FallDown - RiseUp - RiseLeft - RiseRight - RiseDown - Random default: fade description: >- Visual transition/effect type used for each scene. 'Random' picks a dynamic effect per scene. floating: type: boolean default: true description: >- If true, images in each scene gently float or zoom, adding dynamic movement. MusicConfig: type: object description: >- Specifies a background music track and how it should be mixed with voiceover. properties: enabled: type: boolean default: false description: If true, a background music track is included behind the voiceover. musicSlug: type: string enum: - '' - temple_of_treasures - gentle_ambient_loop - serene_ambience - soothing_ambience - soothing_ambient_backdrop - tranquil_ambience - cozy_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 - news_report_theme - adventurous_intro - burlesque_sweetheart - highway_nocturne_national_sweetheart - haptic_sensation - heart_strings_palace - sada_trio - mozart_piano_concerto_no23 - wind_on_our_side - shadows_in_the_sun - native_dreamscape - holly_woods_legend - sky_king_theme - cloud_gentle_ambience - majestic_requiem - contemplative_ambient_intro - quirky_slacker_theme default: '' description: >- The internal identifier of the music track. Must match a known track on the server. You can listen to all the songs on [this page](https://longstories.ai/resources/music-tracks). volume: type: number minimum: 0 maximum: 1 default: 0.3 description: >- Relative loudness of the music track compared to the voiceover. 0 = silent, 1 = full. loop: type: boolean default: true description: Whether the background track automatically loops when it ends. VideoConfig: type: object description: >- Configures video generation for each scene using an image-to-video model. properties: enabled: type: boolean default: false description: >- If true, attempts to generate a video clip for each scene instead of just using the static image (longer render times, higher cost). model: type: string enum: - ltx_video_13b_distilled - ltx_video - seedance_480p_5s - stable_video - motion - seedance_480p_10s - seedance_720p_5s - kling_v1_6_std_5s - kling_v2_1_std_5s - ltx_video_13b_dev - seedance_720p_10s - 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 - kling_v2_1_master_10s - veo2 - kling_v2_master_10s default: luma_ray2 description: >- Which image-to-video model to use. Affects style, quality, duration limits, and cost. See [Video Models](https://longstories.ai/resources/video-models) for details. videoDirectorNotes: type: string description: >- Optional notes specific to the video generation model, potentially influencing motion or style. Ignored by some models. nullable: true duration: type: number description: >- Target duration in seconds for the generated video clip. Some models have fixed durations. Optional. nullable: true securitySchemes: ApiKeyAuth: type: apiKey in: header name: x-api-key description: >- Your API key for authenticating requests. Example: 'x-api-key: MY_SECRET_KEY' ````