> ## Documentation Index > Fetch the complete documentation index at: https://docs.aihubmax.com/llms.txt > Use this file to discover all available pages before exploring further. # Gemini Format - Music Generation > - Uses the Gemini native format generateContent endpoint to generate music via the Lyria 3 model - Enable audio output by including `AUDIO` in `generationConfig.responseModalities`; if `TEXT` is also included, the response will additionally return text descriptions (lyrics/structure) - Supports text prompts and image input (up to 10 images); images are used to inspire visually-driven music creation - Duration, structure (verse/chorus/bridge), style, etc. are primarily controlled via text prompts - `lyria-3-clip-preview`: generates fixed 30-second clips, returns MP3 by default (`audio/mpeg`) - `lyria-3-pro-preview`: generates full songs; you can request `audio/mpeg` or `audio/wav` via `responseMimeType`, but the actual output format should be determined by the `inlineData.mimeType` in the response - For SSE streaming output, use `/v1beta/models/{model}:streamGenerateContent?alt=sse` - Music generation is a single-turn process; multi-turn iterative editing is not supported ## OpenAPI ````yaml openapi/en/gemini-format-music-generation.json POST /v1beta/models/{model}:generateContent openapi: 3.1.0 info: title: Gemini Format - Music Generation version: '1.0' servers: - url: https://api.aihubmax.com security: - BearerAuth: [] paths: /v1beta/models/{model}:generateContent: post: tags: - Text Series > Gemini Format summary: Gemini Format - Music Generation description: > - Uses the Gemini native format generateContent endpoint to generate music via the Lyria 3 model - Enable audio output by including `AUDIO` in `generationConfig.responseModalities`; if `TEXT` is also included, the response will additionally return text descriptions (lyrics/structure) - Supports text prompts and image input (up to 10 images); images are used to inspire visually-driven music creation - Duration, structure (verse/chorus/bridge), style, etc. are primarily controlled via text prompts - `lyria-3-clip-preview`: generates fixed 30-second clips, returns MP3 by default (`audio/mpeg`) - `lyria-3-pro-preview`: generates full songs; you can request `audio/mpeg` or `audio/wav` via `responseMimeType`, but the actual output format should be determined by the `inlineData.mimeType` in the response - For SSE streaming output, use `/v1beta/models/{model}:streamGenerateContent?alt=sse` - Music generation is a single-turn process; multi-turn iterative editing is not supported operationId: gemini-format-music-generation parameters: - name: model in: path required: true schema: type: string example: lyria-3-clip-preview enum: - lyria-3-clip-preview - lyria-3-pro-preview description: >- Model name. `lyria-3-clip-preview` generates 30-second clips (default MP3 / `audio/mpeg`). `lyria-3-pro-preview` generates full songs; you can request `audio/mpeg` or `audio/wav`, but the actual output format should be determined by the returned `inlineData.mimeType` requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/MusicGenerationRequest' examples: text-to-music: summary: Text-to-music (basic) description: >- Generate a music clip from a text prompt, returns audio + text (lyrics/structure description) value: contents: - role: user parts: - text: >- Create a 30-second upbeat folk song using guitar and harmonica. generationConfig: responseModalities: - AUDIO - TEXT full-song: summary: Full song (with structure) description: >- Use Lyria 3 Pro to generate a full song with verse/chorus structure (change the model in the path to lyria-3-pro-preview) value: contents: - role: user parts: - text: >- Create a 3-minute indie pop song in C major, 120 BPM. [Intro] Soft piano arpeggios [Verse 1] Gentle vocals with acoustic guitar [Chorus] Full band arrangement, drums and bass join in, uplifting melody [Verse 2] Add subtle synth layers [Chorus] Repeat, add harmonies [Outro] Piano fade out generationConfig: responseModalities: - AUDIO - TEXT wav-output: summary: WAV format output (Pro only) description: >- Use Lyria 3 Pro to generate music in WAV format (change the model in the path to lyria-3-pro-preview) value: contents: - role: user parts: - text: >- An atmospheric ambient track with ethereal synth pads and gentle rain sounds. generationConfig: responseModalities: - AUDIO - TEXT responseMimeType: audio/wav image-to-music: summary: Image-driven music (image-to-music) description: >- Upload an image + text description; the model generates music based on the image's colors, mood, and atmosphere value: contents: - role: user parts: - text: >- Based on the colors and atmosphere of this image, create an atmospheric ambient piece. - inlineData: mimeType: image/jpeg data: BASE64_ENCODED_IMAGE_DATA generationConfig: responseModalities: - AUDIO - TEXT responses: '200': $ref: '#/components/responses/MusicGenerationSuccess' '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' '402': $ref: '#/components/responses/PaymentRequired' '403': $ref: '#/components/responses/Forbidden' '429': $ref: '#/components/responses/TooManyRequests' '500': $ref: '#/components/responses/InternalServerError' components: schemas: MusicGenerationRequest: type: object required: - contents - generationConfig properties: contents: type: array description: >- Content list. Music generation is a single-turn process; multi-turn iterative editing is not supported items: type: object required: - role - parts properties: role: type: string description: Role of the message sender enum: - user - model parts: type: array description: >- List of content parts. Supports text and images (up to 10); images are used for visually-inspired music creation items: type: object properties: text: type: string description: >- Text content (music generation prompt describing style, instruments, mood, BPM, structure, lyrics, etc.) inlineData: type: object description: >- Image inline data (base64 encoded). The model generates music based on the image's colors, mood, and atmosphere. Supports up to 10 images. required: - mimeType - data properties: mimeType: type: string description: Image MIME type enum: - image/jpeg - image/png data: type: string description: Base64 encoded image data systemInstruction: type: object description: >- System instruction. Lyria 3 model support for this field is not confirmed by official documentation; it may not take effect properties: parts: type: array description: System instruction content parts items: type: object properties: text: type: string description: System instruction text generationConfig: type: object description: >- Generation config; `responseModalities` must include `AUDIO` for music generation requests required: - responseModalities properties: responseModalities: type: array description: >- Response modalities. Must include `AUDIO` for music generation requests; if `TEXT` is also included, text descriptions will be returned additionally items: type: string enum: - AUDIO - TEXT example: - AUDIO - TEXT responseMimeType: type: string description: >- Requested output audio format. Recommended `audio/mpeg` or `audio/wav`; currently known that even when `audio/wav` is requested, the actual output may be `audio/mpeg` -- please refer to the `inlineData.mimeType` in the response enum: - audio/mpeg - audio/mp3 - audio/wav temperature: type: number description: >- Sampling temperature. Not used in official Lyria 3 examples; effectiveness not confirmed minimum: 0 maximum: 2 safetySettings: type: array description: Content safety filter settings items: type: object properties: category: type: string description: Harm category enum: - HARM_CATEGORY_HARASSMENT - HARM_CATEGORY_HATE_SPEECH - HARM_CATEGORY_SEXUALLY_EXPLICIT - HARM_CATEGORY_DANGEROUS_CONTENT threshold: type: string description: Blocking threshold enum: - BLOCK_NONE - BLOCK_LOW_AND_ABOVE - BLOCK_MEDIUM_AND_ABOVE - BLOCK_ONLY_HIGH MusicGenerationResponse: type: object properties: candidates: type: array description: List of generation result candidates items: type: object properties: content: type: object description: Generated content properties: parts: type: array description: >- Content parts, may contain text (lyrics/structure description) and audio; the actual audio MIME type should be determined by `inlineData.mimeType` items: type: object properties: text: type: string description: >- Generated text content (lyrics, song structure description) example: >- Here is an upbeat folk song featuring guitar and harmonica. inlineData: type: object description: Generated audio data properties: mimeType: type: string description: >- Audio MIME type; actual return is typically `audio/mpeg` or `audio/wav` example: audio/mpeg data: type: string description: Base64 encoded audio data example: SUQzBAAAAAAAI1RTU0UAAAAPAAADTGF2ZjU4... role: type: string description: Response role enum: - model finishReason: type: string description: Reason for generation stop enum: - STOP - MAX_TOKENS - SAFETY - RECITATION - OTHER example: STOP safetyRatings: type: array description: Safety ratings for the generated content items: type: object properties: category: type: string description: Harm category probability: type: string description: Probability level usageMetadata: type: object description: Token usage statistics properties: promptTokenCount: type: integer description: Prompt token count example: 15 candidatesTokenCount: type: integer description: Candidates token count example: 200 totalTokenCount: type: integer description: Total token count example: 215 ErrorResponse400: type: object properties: error: type: object properties: message: type: string example: Invalid request parameters type: type: string description: Error type; not all errors return this field example: http_error code: description: Error code; may be an HTTP numeric code or a string error code oneOf: - type: integer example: 400 - type: string example: http_400 status: type: string description: Gemini-style status code; not all errors return this field example: INVALID_ARGUMENT ErrorResponse401: type: object properties: error: type: object properties: message: type: string example: Invalid API Key type: type: string description: Error type; not all errors return this field example: authentication_error code: description: Error code; may be an HTTP numeric code or a string error code oneOf: - type: integer example: 401 - type: string example: http_401 status: type: string description: Gemini-style status code; not all errors return this field example: UNAUTHENTICATED ErrorResponse403: type: object properties: error: type: object properties: message: type: string example: Access denied type: type: string description: Error type; not all errors return this field example: permission_error code: description: Error code; may be an HTTP numeric code or a string error code oneOf: - type: integer example: 403 - type: string example: http_403 status: type: string description: Gemini-style status code; not all errors return this field example: PERMISSION_DENIED ErrorResponse429: type: object properties: error: type: object properties: message: type: string example: Rate limit exceeded type: type: string description: Error type; not all errors return this field example: rate_limit_error code: description: Error code; may be an HTTP numeric code or a string error code oneOf: - type: integer example: 429 - type: string example: http_429 status: type: string description: Gemini-style status code; not all errors return this field example: RESOURCE_EXHAUSTED ErrorResponse500: type: object properties: error: type: object properties: message: type: string example: Internal server error type: type: string description: Error type; not all errors return this field example: server_error code: description: Error code; may be an HTTP numeric code or a string error code oneOf: - type: integer example: 500 - type: string example: http_500 status: type: string description: Gemini-style status code; not all errors return this field example: UNKNOWN responses: MusicGenerationSuccess: description: Music generation response content: application/json: schema: $ref: '#/components/schemas/MusicGenerationResponse' BadRequest: description: Bad request content: application/json: schema: $ref: '#/components/schemas/ErrorResponse400' Unauthorized: description: Unauthorized content: application/json: schema: $ref: '#/components/schemas/ErrorResponse401' PaymentRequired: description: Insufficient balance content: application/json: schema: $ref: '#/components/schemas/ErrorResponse400' Forbidden: description: Forbidden content: application/json: schema: $ref: '#/components/schemas/ErrorResponse403' TooManyRequests: description: Too many requests content: application/json: schema: $ref: '#/components/schemas/ErrorResponse429' InternalServerError: description: Internal server error content: application/json: schema: $ref: '#/components/schemas/ErrorResponse500' securitySchemes: BearerAuth: type: http scheme: bearer description: | ## All endpoints require Bearer Token authentication ## Add the following to your request headers: `Authorization: Bearer YOUR_API_KEY` ````