Seedance 2.5 Flash now 35% off — ends Nov 30
Hero background 1Hero background 2

ElevenLabs Music v2.5 Complete Song Generation

ElevenLabs Music v2.5 is a music generation family from ElevenLabs that creates songs and instrumentals across genres and languages from text prompts. Choose the Detailed endpoint for track titles, genres and optional per-word lyric timestamps alongside the audio. Access both endpoints on Atlas Cloud at a standard price of $0.188 per minute. Start building today.

ElevenLabs Music v2.5 is developed by ElevenLabs. Atlas Cloud (operated by Atlas Cloud AI LLC) provides access to it and does not own it. All trademarks belong to their respective owners.

Explore the Leading ElevenLabs Music v2.5

Atlas Cloud provides you with the latest industry-leading creative models.

ElevenLabs Music v2.5: Compare Music Generation Outputs

Choose song generation alone or add composition metadata and optional lyric timestamps to your audio workflow.

ModalityDescription
ElevenLabs Music v2.5 Text-to-MusicTurn a text prompt into a song or instrumental lasting from 3 seconds to 10 minutes. Support for any genre and many languages makes this endpoint suitable for background scores, multilingual tracks, and complete songs.
ElevenLabs Music v2.5 Text-to-Music (Detailed)When your application needs both music and descriptive data, this endpoint generates a song from a text prompt and returns its title, genres, and section-by-section composition plan. Optional per-word lyric timestamps support workflows that display lyrics alongside audio playback.

ElevenLabs Music v2.5: From Musical Ideas to Structured Songs

Shape ElevenLabs Music v2.5 songs through flexible durations, vocal or instrumental generation, section-level composition controls, and detailed metadata, with transparent pricing on Atlas Cloud.

ElevenLabs Music v2.5: 3 Seconds to 10 Minutes

ElevenLabs Music v2.5 generates songs and instrumentals from text, with durations ranging from 3 seconds to 10 minutes. Describe the genre, mood, and musical direction, then choose a length that fits your project. A single brief can guide anything from a short musical cue to a full composition. This flexibility suits soundtrack sketches, creator content, and song demos.

Vocal Songs or Pure Instrumentals

Choose vocal music in many languages or enable force_instrumental for a track without vocals. In prompt mode, describe the song you want and specify the intended language when requesting singing. Instrumental mode keeps the result focused on music rather than a vocal performance. Use these options to develop song concepts or create background arrangements that leave room for narration.

Arrange ElevenLabs Music v2.5 by Section

For structured songwriting, composition_plan replaces the text prompt with sections containing lyrics, styles to include or avoid, and durations. Define how each part should sound instead of placing every instruction in one brief. With plan mode, seed can help make repeated requests more consistent, although exact reproduction is not guaranteed. These controls suit arrangement drafts that need deliberate structure and pacing.

Song Metadata and Word Timestamps

The Detailed endpoint returns generated music with its title, genres, and composition plan, plus optional timestamps for individual lyric words. Enable with_timestamps when your workflow needs timing information alongside the song. Keep the returned plan with the audio to inspect how its sections are organized. These additional outputs help developers build track catalogs, lyric displays, and editing workflows around generated music.

Transparent Pricing Across Both Endpoints

Atlas Cloud lists a standard price of $0.188 per minute for both the standard and Detailed music endpoints. Choose the standard endpoint for music generation or the Detailed version when you also need metadata. Both use an asynchronous task workflow, with submission followed by prediction polling. This shared request pattern helps developers integrate music generation while planning costs around the duration they request.

ElevenLabs Music v2.5 in the Mix: One Prompt, Three Models

Compare ElevenLabs Music v2.5 with two music models using identical prompts focused on arrangement, instrumental detail, and dynamic transitions.

Prompt

Compose an instrumental synthwave soundtrack for a midnight drive that ends at sunrise. Open with a delicate analog arpeggio and warm sustained chords, leaving space between phrases. Introduce a pulsing bass line, then build tension with increasingly active percussion. Briefly strip back to the arpeggio before a bright, expansive drop with punchy drums and a memorable lead melody. Bring the opening motif back in a softer closing passage and resolve with a clear final chord. Keep the melody recognizable throughout. No vocals, spoken words, or environmental sound effects.

Generated with ElevenLabs Music v2.5 Text-to-Music on Atlas Cloud

Generated with MiniMax Music 3.0 on Atlas Cloud

Generated with MiniMax Music 2.6 on Atlas Cloud

Prompt

Write an instrumental chamber waltz for a curious detective exploring an abandoned toy shop. A solo music box introduces a playful but slightly uneasy melody. Answer it with pizzicato cello and a winding clarinet phrase, keeping the triple meter clear. As the mystery deepens, add bowed strings and darker harmonies without losing the original tune. Let the ensemble fall away for a fragile music box reprise, then close with a surprising, gently reassuring resolution. Preserve distinct instrumental textures and expressive changes in intensity. No vocals, narration, or environmental sound effects.

Generated with ElevenLabs Music v2.5 Text-to-Music on Atlas Cloud

Generated with MiniMax Music 3.0 on Atlas Cloud

Generated with MiniMax Music 2.6 on Atlas Cloud

ElevenLabs Music v2.5 for Songs, Scores, and Stories

Whether you are drafting multilingual songs, scoring videos, podcasts, or games, or preparing lyric videos, ElevenLabs Music v2.5 turns text prompts into music, with optional lyric timestamps available through the detailed endpoint.

Songwriting Demos with ElevenLabs Music v2.5

Turn a text prompt into a complete vocal song or instrumental in your chosen genre. Songwriters can use these drafts to explore musical directions before developing a fuller production brief.

Video and Ad Scores

For video projects, generate instrumental music from text prompts with durations from 3 seconds to 10 minutes. Use the tracks as soundtrack candidates for product videos, ads, and social clips.

Podcast Cues with ElevenLabs Music v2.5

A short text description can become an instrumental cue for a podcast introduction or transition. Give editors musical options for opening segments, separating chapters, and closing episodes around their spoken content.

Game Scene Soundtracks

When a game scene needs music, describe its mood and genre to generate an instrumental track. Indie developers can audition background scores for menus, exploration sequences, or narrative prototypes before final audio integration.

Multilingual Songs with ElevenLabs Music v2.5

Songs with vocals can be generated in many languages from text prompts across musical genres. This gives creators a way to explore song concepts for different audiences through distinct language and style choices.

Lyric Video Production

The detailed endpoint pairs generated songs with optional word timestamps for their lyrics, alongside titles and genres. Video editors and application developers can use that timing data to prepare lyric overlays.

ElevenLabs Music v2.5 and MiniMax: Duration, Control, and Cost

Compare ElevenLabs Music v2.5 with MiniMax music models by track length, instrumental support, arrangement controls, and standard Atlas Cloud pricing.

ModelTrack DurationInstrumental ModeArrangement ControlsStandard Price
ElevenLabs Music v2.5 Text-to-Music3 seconds to 10 minutes√Composition plan with section lyrics, styles, and durations$0.188/minute
ElevenLabs Music v2.5 Text-to-Music (Detailed)3 seconds to 10 minutes√Section composition plan, plus returned metadata and optional lyric timestamps$0.188/minute
MiniMax Music 3.0Up to 5 minutes√Style prompt and lyric section tags$0.15/generation
MiniMax Music 2.6Up to approximately 5 minutes√Lyrics organized with arrangement tags and pause controls-

How to Use ElevenLabs Music v2.5 on Atlas Cloud

Get started in minutes — follow these simple steps to integrate and deploy models through Atlas Cloud's platform.

Create an Atlas Cloud Account

Sign up at atlascloud.ai and complete verification. New users receive free credits to explore the platform and test models.

Why Use ElevenLabs Music v2.5 on Atlas Cloud

Combining the advanced ElevenLabs Music v2.5 models with Atlas Cloud's GPU-accelerated platform provides unmatched performance, scalability, and developer experience.

Performance & flexibility

Low Latency:
GPU-optimized inference for real-time reasoning.

Unified API:
Run ElevenLabs Music v2.5, GPT, Gemini, and DeepSeek with one integration.

Transparent Pricing:
Predictable per-token billing with serverless options.

Enterprise & Scale

Developer Experience:
SDKs, analytics, fine-tuning tools, and templates.

Reliability:
99.99% uptime, RBAC, and compliance-ready logging.

Security & Compliance:
SOC 2 Type II, HIPAA alignment, data sovereignty in US.

ElevenLabs Music v2.5: From First Prompt to Composition Control

ElevenLabs Music v2.5 is an AI music generation model developed by ElevenLabs that creates songs and instrumentals from text descriptions. Atlas Cloud provides standard and Detailed endpoints for integrating music generation into applications.

Both vocal songs and instrumental tracks are supported. In prompt mode, set force_instrumental to true when you need music without vocals. Leaving it false allows the prompt to guide whether the result includes singing.

Choose elevenlabs/music-v2.5/text-to-music or elevenlabs/music-v2.5/text-to-music-detailed and submit an asynchronous generation request using your Atlas Cloud API key. Use the returned prediction ID to poll GET /api/v1/model/prediction/{id} until the task finishes, then retrieve the generated output.

Atlas Cloud lists a standard price of $0.188 per minute for both the standard and Detailed endpoints. Generation is billed by audio duration, so the rate is the same whether you choose audio generation alone or the endpoint with additional metadata.

Supported track lengths range from 3 seconds to 10 minutes. For prompt-based generation, music_length_ms specifies the requested duration in milliseconds. With a composition_plan, individual section durations define the arrangement.

The model supports music generation across genres and vocal output in many languages. Describe the genre, mood, instruments, and desired vocal language in your prompt. When using a composition plan, write style instructions in English while providing lyrics in your chosen language.

Use composition_plan when you need control over individual sections rather than a single overall description. Specify lyrics, positive_styles, negative_styles, and duration for each section to shape the arrangement. A composition_plan replaces prompt, so do not send both in the same request.

Enable with_timestamps on the Detailed endpoint to request timestamps for individual words in the generated lyrics. These timing values can help align lyric displays or captions with the generated song.

Choose the standard endpoint when your application primarily needs generated music. The Detailed endpoint also returns metadata such as the title, genres, and section-by-section composition plan, with optional word-level lyric timestamps. Both endpoints have the same standard price on Atlas Cloud.

Reusing a seed with the same composition_plan and generation parameters can improve consistency, but exact reproducibility is not guaranteed. The seed parameter is available with composition_plan and cannot be combined with prompt. Outputs may also change after system updates.

Explore More Families

Seedance 2.5

The Seedance 2.5 API gives developers ByteDance Seed’s joint audio and video model, generating 4 to 30 seconds in one pass at up to native 4K from text, a first frame, or up to 50 references across images, video, and audio. It edits selected regions, extends clips, and follows white-model and green-screen inputs.

View Family

Seedance 2.5 Flash

The Seedance 2.5 Flash API gives developers the faster, lower-cost tier of ByteDance Seed’s Seedance 2.5, released on November 12, 2026. It generates video at nearly twice the speed of Seedance 2.5, renders up to 30 seconds per clip at 480p or 720p, and takes up to 10 images, 5 videos, and 5 audio clips per request, with video editing for quick variants.

View Family

Kling 4.0

The Kling 4 API delivers Kling 4.0, the video model Kuaishou released on October 15, 2026. It generates up to 30 seconds in a single pass, follows up to 10 keyframes, blends 15 images, videos, and subjects through Omni Reference, and adds stereo audio with lip sync. Atlas Cloud serves all four endpoints, including video editing, on one OpenAI-compatible key from $0.11 per second. Start building today.

View Family

Seedream 5.5

Seedream 5.5 is ByteDance Seed's image generation and editing family. Guide edits with up to 10 reference images, or split an image into a base image and up to 99 editable layers, with optional OCR for text layers. On Atlas Cloud, generation, editing, and layer decomposition each cost a standard $0.02 per image, helping developers plan image workflow budgets. Start building today.

View Family

Kling 4.0 Flash

As a lower-cost option in the Kling 4.0 family, Kling 4.0 Flash combines images, videos, and subjects to create consistent 720p clips lasting 3 to 20 seconds. Start from a text prompt or animate a first-frame image to suit different production workflows. Atlas Cloud offers all four model variants at a standard price of $0.094 per second. Start building today.

View Family

Wan 3.0

Wan 3.0 is Alibaba’s newest video model, producing up to 30 seconds of 1080p video at 30fps with audio in one generation. Steer it with a text prompt, a first frame, up to 10 reference images, or a document. Standard costs less per second, while Prime keeps the same capabilities and returns results faster. Both run on the same Atlas Cloud key as Wan 2.7, 2.6, and 2.5.

View Family

GPT Image 2.6

The GPT Image 2.6 API gives developers OpenAI’s image model released on November 4, 2026, in two versions: Flare for fast iteration and Sunburst for detailed visual work. It generates images 50% faster than before and adds animated GIF output, SVG export, markup edits that target the area you circle, and a complexity setting. Atlas Cloud serves both models on one OpenAI-compatible key with pay-as-you-go pricing.

View Family

Kling Image 4.0

Kling Image 4.0 is Kling AI's image generation model for detailed visual design. Reference images guide compositions, while long layouts bring product imagery and brand messaging into complete page visuals. High resolution output provides room for cropping and display. Atlas Cloud simplifies integration with a unified API and transparent pay-as-you-go pricing. Start building today.

View Family

MiniMax H3 Image

MiniMax H3-Image builds on MiniMax's H3 family, combining image generation and editing in one checkpoint. Detailed prompts guide characters, scenes, and composition, while reference inputs steer visual changes. Atlas Cloud offers hosted access with pay-as-you-go billing, helping you integrate image workflows without managing model infrastructure. Start building today.

View Family

MiniMax H3

MiniMax H3 is MiniMax's multimodal video family for text, image, and reference guided creation. Across supported routes, it preserves subjects from reference media, offers flexible aspect ratios, and pairs generated sound with visuals through H3 Developer, with output profiles selected by endpoint. Atlas Cloud unifies the family behind one OpenAI-compatible key with transparent pay-as-you-go pricing from the standard rate of $0.038 per second. Start building today.

View Family

GPT Image 2.5

GPT Image 2.5 is OpenAI’s image generation and editing model, released on September 8, 2026 in two API versions. Flare is the default, with up to 50% lower latency than GPT Image 2, while Sunburst takes longer to give finer control over detailed edits. Both support up to 16 reference images, five quality tiers, transparent backgrounds, and custom sizes up to 3840px. Atlas Cloud serves the GPT Image 2.5 API at OpenAI’s per token rates, or at a flat price per image by 1K, 2K, or 4K tier.

View Family

Seedream 5.0 Pro

Seedream 5.0 Pro API gives developers ByteDance's controllable image editing model on Atlas Cloud. It places edits precisely with anchors and coordinates, separates images into editable layers, fuses multiple references, and matches exact colors and materials, with multilingual text at 2K and 3K. On Atlas Cloud you reach it through one key!

View Family

One API for All Media AI.

Explore all models