can-i-run-this-llm

Rig report

NVIDIA RTX 5090 (32GB)

32.0 GB memory · 1792 GB/s · 35 of 48 tracked models fit

Best model it can run

Gemma 4 31BRuns great

Q5_K_M · 21.7 GB · 50–75 tok/s · capability 84/100

Fastest that fits: Kimi Linear 48B-A3B (MoE) at 263–395 tok/s

Also runs

Open the full calculator →Check another machine

Compare against

vs MacBook Pro M4 Max (128GB) →vs NVIDIA DGX Spark (GB10, 128GB) →vs MacBook Pro M5 Max (128GB) →vs MacBook Pro M5 Max (36GB) →vs Mac Studio M3 Ultra (96GB) →vs MacBook Pro M5 Pro (64GB) →vs MacBook Pro M4 Max (48GB) →vs AMD Ryzen AI Max+ 395 “Strix Halo” (128GB) →all comparisons →

Have a NVIDIA RTX 5090 (32GB)? Add your real benchmark → to sharpen these numbers for everyone.

Estimates from device specs and model architectures, refined by community benchmarks. See the methodology.