myrtle.ai's VOLLO has set a new STAC-ML record in gradient boosting tree inference.
In the audited benchmark tests announced today at STAC Summit in London, the 99th percentile latency is below 2 microseconds, with a throughput of up to 50 million inferences per second.
Cambridge, UK, October 6, 2026 / PRNewswire / -- Today, it was announced that its VOLLO ® inference accelerator has set a new record in gradient boosting tree tests on the STAC - ML Markets ( Inference ) benchmark, reducing the 99th percentile latency by over 30% and increasing throughput by at least 5 times. The results, after STAC ® auditing, were announced today at the STAC Summit held in London.
VOLLO runs on the AMD Alveo ™ V80LL computing accelerator within the Blackcore token issuance N 3132- SM + servers, achieving a p99 latency of less than 2 microseconds across all three models. For the smallest model, it continuously performed 50 million inferences per second with a p99 latency of only 1.77 microseconds.
In electronic trading, the time between the arrival of market data and the making of decisions directly affects profits. Low and predictable delays enable institutions to operate larger, more accurate models without missing market opportunities; therefore, the quality of these models no longer has to compromise for speed.
myrtle.ai CEO Peter Baldwin stated: "Trading companies want to run increasingly powerful models without sacrificing speed, and these results show that they can achieve this. Developers now have the ability to test their models on VOLLO without any FPGA experience and see the differences for themselves."
Following the release of the results for STAC Tacana in April, VOLLO now holds the record for both deterministic latency in decision trees and neural networks. This product has been verified in a production environment, and hundreds of thousands of hours of actual trading have generated alpha for many leading trading companies around the world. Its model flexibility also makes it the preferred platform in the fields of telecommunications, network security, and national defense.
STAC - ML Markets ( Inference ) is a technical benchmark standard designed for real-time market data inference operations. This benchmark was developed by quantitative and technical personnel from leading financial institutions and is used to evaluate the performance, resource efficiency, and quality of any technical stack that can run the provided models. For the complete results, please refer to the STAC report ( SUT ID MRTL2026905 ), available at the URL www.STACresearch.com / MRTL2026905.
ML Developers now have the ability to evaluate the performance of their models on VOLLO without the need for FPGA tools or related experience. Please visit myrtle.ai / vollo-trees or contact [ email protected ].
About myrtle.ai
myrtle.ai is a AI / ML software company that provides ultra-low latency inference accelerators on platforms based on FPGA from major leading FPGA suppliers. Its accelerators are applied in scenarios such as financial transactions, wireless telecommunications, large language models, speech processing, and recommendation systems.
The VOLLO, VOLLO Accelerator, and VOLLO designations are registered trademarks of myrtle.ai. “STAC” and all names that start with STAC are trademarks or registered trademarks of Strategic Technology Analysis Center and LLC. The AMD and AMD arrow symbols, Alveo, and their combinations are all trademarks of Advanced Micro Devices.
