boltbeam.fyi

2× Radeon R9700 for local LLMs

4 measured runs. 32 GB of VRAM; memory measured at 0 GB/s, which sets the roofline for every model below.

ModelQuantEnginetok/sRooflineShare
Qwen3.8-27BQuark W4A16vLLM + MTP k=292.3––
Qwen3.8-27BQuark W4A16vLLM56.1––
Qwen3.8-Flash-NextGSQ-RCO IQ3_Sllama.cpp + MTP k=372.5––
Qwen3.8-Flash-NextGSQ-RCO IQ3_Sllama.cpp39.4––

Share = measured tok/s over the roofline: the bytes a token reads over this GPU's measured memory speed. Open a quant for the test machine and the per-role and per-kernel breakdown.

Open the explorer on this GPU →