Chinese model Mureka V8 beats Suno V5 and claims a new music genre

The launch happened in a music club, not a conference hall. In January 2026 the Chinese company Kunlun Tiangong presented the Mureka V8 model and turned the press conference into a concert, with music generated by the new system playing behind the video. Promotional materials stressed that in tests with professional musicians Mureka V8 beat Suno V5, considered the leading music model from Silicon Valley, and took first place in that narrow category.

From "it can be generated" to "it can be released"

The most interesting part of the argument is not the sound itself but the stage the models have reached. According to the industry outlet 量子位, earlier AI output stopped at "it can be generated". The new generation moves the bar to "it can be released", meaning a product that meets market standards. The authors of the test checked four dimensions: singing and expression, musical completeness, production quality and market fit. Mureka V8 is said to match its performance technique to the chosen gender of the vocalist, which makes it sound less mechanical. From a single line of lyrics it generates a track close to three minutes long rather than two, as earlier versions did. The declared "Eastern aesthetic" is part of the deal: the model naturally renders Chinese sounds, as heard in a generated New Year song.

Fang Han, chairman and CEO of Kunlun Wanwei, speaks of a mission broader than selling a tool: "We want music to become a form of expression available to everyone." In his view, good AI tracks are becoming a new genre, like jazz, country or rap, and the Mureka platform is to be a lighthouse for the industry. The Chinese lyricist and producer Gao Xiaosong took part in the launch, and people from leading Chinese record labels reported on the event.

The market responds

The enthusiasm of model makers collides with the reaction of the recording industry. In August 2026 the American rapper Dr. Dre publicly admitted he uses AI in his work and compared it to the drum machine and the synthesizer: "It's a whole new creative tool, I accept it and I can't wait to see what comes of it." He added that those now-common devices also met resistance when they appeared. He did not reveal, however, whether he generates whole tracks or only separates stems. Streaming platforms have meanwhile started tightening their rules on "AI slop": Tidal announced that fully generated tracks will not receive royalties, and Spotify introduced a label for artificial vocalists imitating real people. According to data from the platform Deezer, half of the new music reaching the service today is generated by AI.