community pool / model answer

Qwopus3.8-27B-Flash

chat · Jackrong/Qwopus3.8-27B-Flash

21.3 tok/s · model median · 2 runs in the model record.

Answer

15.6tok/s

reportedRyzen AI Max 395 128GB4-bit · 2 runs · decodebest cell in the pool

  1. Ryzen AI Max 395 128GB · unified memory15.6 tok/sreported2 runs · peak VRAM no data yet
Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
Ryzen AI Max 395 128GB4-bit15.6 tok/sreported2,053- GB790llama.cpp

Context tested

The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.