Today we’re unveiling our next-generation AI model. It delivers up to 20% higher accuracy across a wide set of tests, which translates directly into noticeably better results for sync teams, labels and publishers.
Higher accuracy means fewer dead-end searches, clearer recommendations and faster paths to the right track. It also means catalogs perform better. More of the music you already own or represent becomes discoverable in searches that might have overlooked it in the past.
"The new AIMS Search delivers incredibly fast, highly relevant music results using next level AI prompt search. It understands cultural and complex references by interpreting context from a wide dataset, and translates that into precise track recommendations from our catalogs. The result is a tool that feels less like a search engine and more like a true virtual creative music supervisor: accurate, intuitive, and available 24/7."
Solving the hardest part of music search
Most searches begin with someone trying to describe something that’s hard to put into words. A vague mood. A scene outline. A mix of references. A brief with gaps. Search tools usually struggle with that. AIMS was built to handle it.
The new model sharpens that ability. It understands intent with greater nuance, which makes Prompt Search and our other tools feel more natural to use, even when the input isn’t fully formed.
"AIMS’ new model is fantastic. It is innovative, forward-thinking and genuinely moves the needle in terms of simplifying search and discovery. The accuracy of the results blew our minds. We’ve already started building it into our new site."
A step forward for performance and discoverability
AIMS has long been considered the industry standard for AI-powered music search. Raising performance by 20% at this stage required years of research and iteration. The outcome is a model that reveals more relevant tracks with less friction, across catalogs of any size.
"AIMS is already considered the industry standard for music search technology, so improving its performance by twenty percent is a remarkable achievement by our team. Clients notice it right away in the quality and range of results, because the new model now understands search intent with more nuance and reveals even more relevant tracks from the catalog."
For rights holders, this translates directly into stronger visibility for their music and more opportunities for their artists and writers.
Built in house and shaped by real workflows
Like every major AIMS update, this model was developed in house. It took years of research, countless iterations and continuous client feedback. The goal hasn’t changed: make search faster, fairer and more intuitive for both experts and non-experts.
"This model was built in-house, drawing on years of research experience and a team that's always looking for the next improvement. We also learn a lot from our clients, whose feedback keeps us focused on what actually matters. This release is an important milestone, but the journey is far from finished, and we are excited about where we can take our technology next."
Continuing the AIMS mission
AIMS was founded to improve how music is found. Not with more filters or stricter rules, but with technology that understands sound, language and creative intention. This release brings us closer to that mission and sets a higher standard for what search should feel like.
If you’d like to see the new model in action, register for our webinar on December 4.

.png&w=3840&q=100)