The 'MusicGen' model is a text-to-audio converter that is specialized in generating music. It takes in user parameters like prompt (music type such as “lo-fi guitar”), duration, temperature, normalization_strategy, etc., and generates a .wav file output of the specified music. The user can control the output through various settings like top_k, top_p, continuation, model_version, and classifier_free_guidance. The model's output is a URL link to the generated audio file.

Use cases

The musicgen AI model could be used in a number of creative and innovative ways, given its ability to generate custom music based on descriptive text inputs. For instance, video production companies could leverage this model to produce unique background scores for their various content pieces, sidestepping the need for expensive licensing deals. The AI could create customized music, aligning to the feel and theme of the film or video. Similarly, musicians and composers might utilize musicgen to prompt fresh musical ideas, or to compose parts of their work quickly. This could revolutionize the music production process, speeding up composition and allowing musicians to fine-tune their focus on the creative aspects. The musicgen AI model could also be applied in the development of adaptive video games. It could generate different types of music as the game progresses, reacting to the player's choices and creating a more immersive experience. In terms of specific products, a software solution for filmmakers, artists, and video game developers could be created, facilitating the generation of unique soundtracks. Alternatively, a platform could be developed that allows users to type in descriptions of the mood or style of music they want to listen to, and the model would create it on-the-spot, offering a personalized music listening experience. It could even be built into smart speakers or similar devices, providing ambient sounds based upon preferences given vocally by the user.



Cost per run
Avg run time

Creator Models


Similar Models

Try it!

You can use this area to play around with demo applications that incorporate the Musicgen model. These demos are maintained and hosted externally by third-party creators. If you see an error, message me on Twitter.

Currently, there are no demos available for this model.


Summary of this model and related resources.

Model NameMusicgen
Deployment of Meta's MusicGen
Model LinkView on Replicate
API SpecView on Replicate
Github LinkView on Github
Paper LinkNo paper link provided


How popular is this model, by number of runs? How popular is the creator, by the sum of all their runs?

Model Rank
Creator Rank


How much does it cost to run this model? How long, on average, does it take to complete a run?

Cost per Run$-
Prediction Hardware-
Average Completion Time-