Rig report
NVIDIA RTX 4090 (24GB)
24.0 GB memory · 1008 GB/s · 15 of 18 tracked models fit
Best model it can run
Gemma 4 31BTight / slow
Q3_K_M · 13.2 GB · 33–50 tok/s · capability 82/100
Fastest that fits: Kimi Linear 48B-A3B (MoE) at 242–364 tok/s
Have a NVIDIA RTX 4090 (24GB)? Add your real benchmark → to sharpen these numbers for everyone.
Estimates from device specs and model architectures, refined by community benchmarks. See the methodology.