Configs › qwen3-4b on Strix Halo, DGX Spark and Mac: measured tok/s

qwen3-4b on Strix Halo, DGX Spark and Mac: measured tok/s

Every tracked benchmark config for qwen3-4b on local hardware: quant, backend, context, tokens per second, source-linked and trust-tiered.

Snapshot 2026-09-30 · 254 configs · 116 models. Every number is generated from the tracker snapshot and linked to where it was measured; nothing here is typed by hand.

Measured configs for qwen3-4b

ModelQuantBackendDecodeContextTrustSource
qwen3-4bMLX-4BITmlx186 tok/s (prefill 5506)512Extracted, source-linkedsource
qwen3-4bQ4_K_Mllama.cpp159.7 tok/s (prefill 4973)512Extracted, source-linkedsource

More: all configs · hardware · methodology · llms.txt · llms-full.txt · API (OpenAPI) · JSON snapshot. Built by Altronis, private on-prem AI, Singapore.