can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX 5080 (16GB) run Ornith 1.0 35B (MoE)?

Ornith · MIT · estimated for a 16.0 GB machine at 4K context.

Yes, but it's tight.

Best fit: Ornith 1.0 35B (MoE) at Q2_K · 12.3 GB · 231–346 tok/s on a NVIDIA RTX 5080 (16GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K12.3 GB231–346 tok/sTight / slow
Q3_K_M16.7 GB205–307 tok/sWon't fit
Q4_K_M22.1 GB180–271 tok/sWon't fit
Q5_K_M26.5 GB165–247 tok/sWon't fit
Q6_K29.3 GB156–234 tok/sWon't fit
Q8_036.9 GB136–204 tok/sWon't fit
FP1669.4 GB88–133 tok/sWon't fit
Open the full calculator →Everything a NVIDIA RTX 5080 (16GB) runsOrnith 1.0 35B (MoE) on other machinesWhat to buy to run it →

Ran Ornith 1.0 35B (MoE) on a NVIDIA RTX 5080 (16GB)? Add your real numbers →