Get a weekly rundown of the latest AI models and research... subscribe!



AI model preview image
musicgen is a text-to-audio model that is capable of generating music based on a given prompt or melody. It uses artificial intelligence algorithms to analyze and understand the musical attributes of the input and then generates a composition that is in line with the style and genre specified. The model allows users to input text prompts or musical notation and receive a corresponding music composition as an output. This can be useful for musicians, composers, and music enthusiasts who are looking for inspiration or want to explore new musical ideas.

Use cases

musicgen can be used in a variety of contexts. Musicians and composers can use it as a creative tool, generating new musical ideas based on the prompts or melodies they provide. It can also be used as a learning tool for music students, helping them to analyze and understand the elements of a specific musical genre or style. Additionally, musicgen can be integrated into music production software or applications, providing users with a quick and easy way to generate background music or fill in missing sections of a composition. Overall, musicgen offers a unique and innovative approach to music creation and exploration. With further development, it has the potential to revolutionize the way music is composed and experienced. Possible products or practical uses of this model could include a standalone application for musicians and composers to generate music on the go, a plugin for popular music production software, or an online platform where users can collaborate and share generated compositions.



Cost per run
Avg run time
Nvidia A100 (40GB) GPU

Creator Models

Zephyr 7b Alpha$?6,826
Sql Generator$?3,556
Falcon 40b Instruct$?37,574
Instructblip Vicuna13b$0.0138251,676
Mplug Owl$0.018453,411

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
Generate music from a prompt or melody
Model LinkView on Replicate
API SpecView on Replicate
Github LinkView on Github
Paper LinkView on Arxiv


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$0.0943
Prediction HardwareNvidia A100 (40GB) GPU
Average Completion Time41 seconds