can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX 4090 (24GB) run GPT-OSS 20B (MoE)?

GPT-OSS · Apache 2.0 · estimated for a 24.0 GB machine at 4K context.

Yes — it runs great.

Best fit: GPT-OSS 20B (MoE) at FP16 · 13.8 GB · 162–243 tok/s on a NVIDIA RTX 4090 (24GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K11.5 GB177–266 tok/sRuns great
Q3_K_M11.5 GB177–265 tok/sRuns great
Q4_K_M11.6 GB176–264 tok/sRuns great
Q5_K_M11.7 GB175–263 tok/sRuns great
Q6_K12.0 GB173–259 tok/sRuns great
Q8_012.1 GB173–259 tok/sRuns great
FP1613.8 GB162–243 tok/sRuns great
Open the full calculator →Everything a NVIDIA RTX 4090 (24GB) runsGPT-OSS 20B (MoE) on other machinesWhat to buy to run it →

Ran GPT-OSS 20B (MoE) on a NVIDIA RTX 4090 (24GB)? Add your real numbers →