# Private Inference on CIFAR-10 Source: [Private Inference on CIFAR-10]() Collection index: [llms.txt]() Data updated: 2026-04-23T06:48:39.582Z Evaluation of Private Inference latency and runtime performance across various network conditions (WAN, 5G, 4G/LTE, IoT) using the CIFAR-10 dataset. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Private Inference | | Dataset (including any stated split) | CIFAR-10 | | Metric | Runtime (WAN) | | Unit | TIME | | Direction | Lower is better | | Results loaded | 12 of 12 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Runtime (WAN)](): Lower is better; unit: TIME. - [Runtime (5G)](): Lower is better; unit: TIME. - [Runtime (4G/LTE)](): Lower is better; unit: TIME. - [Runtime (IoT)](): Lower is better; unit: TIME. - [Relative Runtime (%)](): Lower is better; unit: PERCENT. - [Runtime (LAN)](): Lower is better; unit: TIME. - [Communication Size (MB)](): Lower is better; unit: SIZE. ## Recorded results | Method | Runtime (WAN) | Result date | Report paper | | --- | --- | --- | --- | | Ours | 2.07 | Not recorded | 68fa7a661e209131cea82f1f | | MD-SONIC | 3.83 | Not recorded | 68fa7a661e209131cea82f1f | | MD-ML | 6.37 | Not recorded | 68fa7a661e209131cea82f1f | | Ours | 8.27 | Not recorded | 68fa7a661e209131cea82f1f | | MD-SONIC | 11.14 | Not recorded | 68fa7a661e209131cea82f1f | | MD-ML | 21.51 | Not recorded | 68fa7a661e209131cea82f1f | | StriaNet | 26.61 | Not recorded | 697c3af8a8b01e76cc8da3f2 | | Vanilla Cheetah | 35.1 | Not recorded | 697c3af8a8b01e76cc8da3f2 | | StriaNet | 42.12 | Not recorded | 697c3af8a8b01e76cc8da3f2 | | Vanilla Cheetah | 50.71 | Not recorded | 697c3af8a8b01e76cc8da3f2 | | StriaNet | 72.67 | Not recorded | 697c3af8a8b01e76cc8da3f2 | | Vanilla Cheetah | 81.46 | Not recorded | 697c3af8a8b01e76cc8da3f2 | ### Ours Runtime (WAN): 2.07. Result ID: 69e9c0ac47f34ce924d055bc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Architecture":"LeNet"}. This record does not assert independent reproduction. ### MD-SONIC Runtime (WAN): 3.83. Result ID: 69e9c0ac47f34ce924d055bd. [Reporting paper]() Recorded evaluation configuration: {"Architecture":"LeNet"}. This record does not assert independent reproduction. ### MD-ML Runtime (WAN): 6.37. Result ID: 69e9c0ac47f34ce924d055be. [Reporting paper]() Recorded evaluation configuration: {"Architecture":"LeNet"}. This record does not assert independent reproduction. ### Ours Runtime (WAN): 8.27. Result ID: 69e9c0ac47f34ce924d055bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Architecture":"AlexNet"}. This record does not assert independent reproduction. ### MD-SONIC Runtime (WAN): 11.14. Result ID: 69e9c0ac47f34ce924d055c0. [Reporting paper]() Recorded evaluation configuration: {"Architecture":"AlexNet"}. This record does not assert independent reproduction. ### MD-ML Runtime (WAN): 21.51. Result ID: 69e9c0ac47f34ce924d055c1. [Reporting paper]() Recorded evaluation configuration: {"Architecture":"AlexNet"}. This record does not assert independent reproduction. ### StriaNet Runtime (WAN): 26.61. Result ID: 69a178b835dfcabb00e7460c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Backbone":"ResNet-50","Server CPU Threads":"16x"}. This record does not assert independent reproduction. ### Vanilla Cheetah Runtime (WAN): 35.1. Result ID: 69a178b835dfcabb00e7460d. [Reporting paper]() Recorded evaluation configuration: {"Model Backbone":"ResNet-50","Server CPU Threads":"16x"}. This record does not assert independent reproduction. ### StriaNet Runtime (WAN): 42.12. Result ID: 69a178b835dfcabb00e7460a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Backbone":"ResNet-50","Server CPU Threads":"8x"}. This record does not assert independent reproduction. ### Vanilla Cheetah Runtime (WAN): 50.71. Result ID: 69a178b835dfcabb00e7460b. [Reporting paper]() Recorded evaluation configuration: {"Model Backbone":"ResNet-50","Server CPU Threads":"8x"}. This record does not assert independent reproduction. ### StriaNet Runtime (WAN): 72.67. Result ID: 69a178b835dfcabb00e74608. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Backbone":"ResNet-50","Server CPU Threads":"4x"}. This record does not assert independent reproduction. ### Vanilla Cheetah Runtime (WAN): 81.46. Result ID: 69a178b835dfcabb00e74609. [Reporting paper]() Recorded evaluation configuration: {"Model Backbone":"ResNet-50","Server CPU Threads":"4x"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.