can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA L40S (48GB) run DeepSeek-R1-Distill 8B?

DeepSeek · Llama 3.1 Community · estimated for a 48.0 GB machine at 4K context.

Yes — it runs great.

Best fit: DeepSeek-R1-Distill 8B at FP16 · 16.1 GB · 35–52 tok/s on a NVIDIA L40S (48GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K3.2 GB117–175 tok/sRuns great
Q3_K_M4.0 GB101–152 tok/sRuns great
Q4_K_M4.9 GB89–133 tok/sRuns great
Q5_K_M5.7 GB80–120 tok/sRuns great
Q6_K6.6 GB72–108 tok/sRuns great
Q8_08.5 GB59–88 tok/sRuns great
FP1616.1 GB35–52 tok/sRuns great

Ran DeepSeek-R1-Distill 8B on a NVIDIA L40S (48GB)? Add your real numbers →