Benchmarking Pocket-Scale Inference

Artificial Analysis, in partnership with Liquid AI, has launched a benchmark suite for pocket-scale AI models that fit within 8 GB of memory after quantization, including KV cache at 8K context, measuring intelligence and inference performance on mobile devices such as the iPhone 17 Pro. The tests cover real-world mobile usage scenarios, with independent validation of Liquid AI's inference measurement process, and results are presented as an Average Score across benchmarks like BFCL, IFBench, GPQA Diamond, and MATH-500.

Benchmarking pocket-scale inference We benchmark small models on mobile phones. Artificial Analysis' testing covers model intelligence on a set of benchmarks chosen to represent real-world mobile device usage, and we partner with Liquid AI to gather real inference data measured on the devices themselves. Note: we have independently validated Liquid AI's inference measurement process. “Small” models are all models that fit inside 8 GB of memory after quantization, including KV cache at 8K context. View all rules and our process in the methodology page /methodology/mobile-device-benchmark-set . Intelligence and Inference Performance Summary Average Score 16K max context vs. End-to-End Generation TimeiPhone 17 Pro Inference Performance End-to-End Generation TimeiPhone 17 Pro Model Intelligence Average Score Mobile Device Benchmark Set, 16K max context iPhone 17 Pro Looking for the Artificial Analysis Intelligence Index scores for these models? The following models have been evaluated on our full index, and their scores are visible on their model pages: Token Efficiency Context Budget Overruns Evaluation Breakdown Mobile Device Benchmark Set Evaluations 16K max context iPhone 17 Pro BFCL Tool calling index subset IFBench Instruction following AA-Omniscience Accuracy Knowledge AA-Omniscience Non-Hallucination Rate 1 - hallucination rate GPQA Diamond Scientific reasoning MATH-500 Quantitative reasoning