Liquid AI this week open-sourced Pipette, a benchmarking platform built to measure foundation model performance on edge devices. Rather than treating on-device behavior as a property of a model alone, Pipette defines the unit of measurement as a full deployment configuration: model, quantization, runtime, and device. According to the release cited by Techub News, the initial dataset covers more than 1,000 combinations spanning models, quantization methods, runtimes, devices, and context settings, across over 30 models. The suite is released under the Apache 2.0 license and includes infrastructure, a public results dataset, a hosted dashboard, and native benchmarking apps for iOS and Android. Liquid AI said the toolkit is designed for a range of users, from independent developers to large OEMs, with use cases including model and quantization selection, hardware procurement validation, and regression testing. Pipette was developed with independent validation firm Artificial Analysis, and the published dataset includes five on-device performance metrics. Early validated results come from devices including a MacBook Pro with M5 Max, iPhone 17 Pro, and Galaxy S26 Ultra.
Liquid AI this week open-sourced Pipette, an open platform for benchmarking foundation models on edge devices, according to Techub News.
Pipette treats on-device behavior as a property of the deployment system rather than an isolated model. Its unit of measurement is a full configuration: model, quantization, runtime, and device. The initial dataset covers more than 1,000 combinations across models, quantization options, runtimes, devices, and context configurations, spanning more than 30 models.
What the release includes
Pipette is released under the Apache 2.0 license. The package includes infrastructure, a public results dataset, a hosted dashboard, and native benchmarking apps for iOS and Android.
Use cases and development partners
Liquid AI said the suite is intended for teams ranging from independent developers to large OEMs. It is positioned for model and quantization selection, hardware procurement validation, and regression testing in on-device AI evaluation.
Liquid AI developed Pipette with independent validation firm Artificial Analysis to support the reliability of the methodology. MarkTechPost said the released dataset includes five on-device performance metrics, and early validated results come from devices including a MacBook Pro with M5 Max, iPhone 17 Pro, and Galaxy S26 Ultra.
This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan. Disclaimer:
The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.
Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.
