can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX Pro 6000 Blackwell (96GB) run Ornith 1.5 35B-A3B (MoE)?

Ornith · MIT · estimated for a 96.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Ornith 1.5 35B-A3B (MoE) at FP16 · 71.1 GB · 136–205 tok/s on a NVIDIA RTX Pro 6000 Blackwell (96GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K13.1 GB275–413 tok/sRuns great
Q3_K_M16.7 GB259–389 tok/sRuns great
Q4_K_M21.9 GB239–358 tok/sRuns great
Q5_K_M25.5 GB226–339 tok/sRuns great
Q6_K30.5 GB211–316 tok/sRuns great
Q8_037.8 GB192–288 tok/sRuns great
FP1671.1 GB136–205 tok/sRuns great
Open the full calculator →Everything a NVIDIA RTX Pro 6000 Blackwell (96GB) runsOrnith 1.5 35B-A3B (MoE) on other machinesWhat to buy to run it →

Ran Ornith 1.5 35B-A3B (MoE) on a NVIDIA RTX Pro 6000 Blackwell (96GB)? Add your real numbers →