Created bymario-alka.dePowered bygodcore.denoob2claw.detricoma.de
Beigesteuert vonMario AlkaPoolside

Laguna-S-2.1-INT4

Performancebenchmark · gemessen am 28.07.2026 01:00

Benchmark-IDrun-20260728-034053-2e86cf
Timebench 3 - Kombi (Prefill + Generation)Runtime: llama.cppQuantisierung: INT4
Generation42,33tok/s
Prefill153,14tok/s
Time to First Token257.193,00ms
Gesamtdauer1.200,00s
Concurrency10parallel