Rig report
NVIDIA RTX 5070 (12GB)
12.0 GB memory · 672 GB/s · 27 of 64 tracked models fit
Best model it can run
Devstral Small 2 24BTight / slow
Q2_K · 8.9 GB · 45–68 tok/s · capability 72/100
Fastest that fits: LFM2.5 8B-A1B (MoE) at 188–282 tok/s
Compare against
Have a NVIDIA RTX 5070 (12GB)? Add your real benchmark → to sharpen these numbers for everyone.
Estimates from device specs and model architectures, refined by community benchmarks. See the methodology.