DiffRhythm AI & API: Instant Full-Length Song Generation!

Create complete, professional-quality songs in seconds! DiffRhythm is the world’s first open source latent diffusion model that generates full length vocal and instrumental tracks from simple text prompts. Describe your vision and let DiffRhythm handle the rhythm, melody, and lyrics.

AI Playground Try our API!Join Discord

DiffRhythm Music Generation Demo

Switch between examples to see different input/output combinations.

Audio Reference

Reference audio

Input Lyrics


  [00:00.52]Abracadabra abracadabra
  [00:03.97]Ha
  [00:04.66]Abracadabra abracadabra
  [00:12.02]Yeah
  [00:15.80]Pay the toll to the angels
  [00:19.08]Drawin' circles in the clouds
  [00:23.31]Keep your mind on the distance
  [00:26.67]When the devil turns around
  [00:30.95]Hold me in your heart tonight
  [00:34.11]In the magic of the dark moonlight
  [00:38.44]Save me from this empty fight
  [00:43.83]In the game of life
  [00:45.84]Like a poem said by a lady in red
  [00:49.45]You hear the last few words of your life
  [00:53.15]With a haunting dance now you're both in a trance
  [00:56.90]It's time to cast your spell on the night
  [01:01.40]Abracadabra ama-ooh-na-na
  [01:04.88]Abracadabra porta-ooh-ga-ga
  [01:08.92]Abracadabra abra-ooh-na-na
  [01:12.30]In her tongue she's sayin'
  [01:14.76]Death or love tonight
  [01:18.61]Abracadabra abracadabra
  [01:22.18]Abracadabra abracadabra
  [01:26.08]Feel the beat under your feet
  [01:27.82]The floor's on fire
  [01:29.90]Abracadabra abracadabra

DiffRhythm Generated Music

Generated music output

Features

End-to-End Full-Length Music: Generate complete songs up to 4 minutes 45 seconds in a single step - no stitching short clips or multi-stage workflows.
Style + Scene-Driven Creation!: Describe moods, genres, or imagery (e.g. 'Jazzy Nightclub Vibe ' or 'Indie Folk Ballad with Acoustic Harmonica ') to shape unique compositions.
Asynchronous API Calls: Our aysnchronous API call structure allows developers to submit tasks and have their program continues on, until a "callback" function is executed.
Instrumental Mode: Craft soundscapes from wild prompts like “Arctic Theremin Storms” - perfect for film scores, game soundtracks, or experimental music.
Pure Vocal Generation: Focus on lyrical storytelling with standalone vocal tracks, ideal for refining lyrics or acapella projects.
Multilingual Music: Seamlessly generate songs in English or Chinese, with natural-sounding vocal phrasing in both languages.
High Concurrency: Experience stable performance even under the most demanding load - our service can automatically scale as per varying peak load, processing high number of jobs concurrently while keeping latencies to the minimum!
10-Second Inference Speed: Leverage non-autoregressive architecture to create songs 100x faster than language model-based alternatives.
Open-Source Freedom: Apache 2.0 license allows commercial use, customization, and integration into your creative tools or apps.
Dynamic Length Control: Adjust song duration on the fly, from 30-second jingles to extended 10-minute compositions (coming soon!).
Song Extension & Remixing: Expand existing tracks or blend styles by extending AI-generated songs with new sections (coming soon!).

Our Pricing Plans

"Pay-as-you-go" Option

For the most updated unit pricing, please
see our pricing page more details!

txt2audio base | 1.35 minutes

$0.02/call

txt2audio full | 4.45 minutes

$0.02/call

Check our Pricing page for more information

Detailed DiffRhythm Pricing To our API Playground!

Frequently asked questions

What is DiffRyhthm?: DiffRhythm is the first open-source latent diffusion model developed by ASLP Lab, designed specifically for end-to-end full-length song generation. It creates complete tracks (up to 285s) with both vocals and instrumental accompaniment in seconds, using only text prompts like lyrics and style descriptions. Unlike traditional multi-stage systems, it combines simplicity, speed, and scalability - making AI powered music creation accessible to everyone.
What is the DiffRhythm API?: DiffRhythm API is created by PiAPI and it lets developers integrate our song generation technology directly into apps, tools, or workflows. It provides programmatic access to the model’s capabilities, enabling features like bulk generation, real-time customization, and seamless scaling for commercial use. Whether you’re building a music production app, a game soundtrack engine, or a creative AI platform, the API handles compute-heavy tasks while you focus on user experience.
What types of music can DiffRhythm generate?: The DiffRhythm API supports virtually any genre or style described in your prompts! Whether you need pop, jazz, electronic, folk, cinematic scores, or experimental soundscapes, it adapts to your creative vision.
What is the Pricing for the music generation?: For our "Pay-as-you-go" option, it costs $0.02 for each generation call for both txt2audio base (1.35 mins) and txt2audio full (4.45mins).
Is there a maximum number of concurrent jobs for the "Pay-as-you-go" Option?: Yes, different subscription plans (Free, Creator, or Pro Plan) will have grant the "Pay-as-you-go" users different number of concurrent jobs, please refer to our pricing for details.
Can I use DiffRhythm songs commercially?: Yes! The Apache 2.0 license permits commercial use, but you must verify originality and disclose AI involvement!
Q: What inputs are required?: Just lyrics (optional) and a style prompt. No instrumental references or melody templates needed!
How do I pay for this API?: Our workspace has integrated Stripe in our payment system, which will allow payments to be made from most major credit card providers.
Do you offer refunds?: No, we do not offer refunds. But when you first sign up for an account on PiAPI‘s Workspace, you are given free credits to try our DiffRhythm (specifically the "Pay-as-you-go" service) before making payments!
How can I get in touch with your team?: Please email us at contact@piapi.ai - we'd love to listen to your feedback and explore potential collaborations!