can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA H100 (80GB) run Qwen3.5 4B?

Qwen · Apache 2.0 · estimated for a 80.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Qwen3.5 4B at FP16 · 8.4 GB · 158–236 tok/s on a NVIDIA H100 (80GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M1.8 GB272–408 tok/sRuns great
Q4_K_M2.6 GB252–378 tok/sRuns great
Q5_K_M3.0 GB241–361 tok/sRuns great
Q6_K3.5 GB230–345 tok/sRuns great
Q8_04.5 GB211–316 tok/sRuns great
FP168.4 GB158–236 tok/sRuns great

Ran Qwen3.5 4B on a NVIDIA H100 (80GB)? Add your real numbers →