Q4_K_M: Smooth Fit 25.03 GB / 78.4 GB Usable
Qwen3-32B on NVIDIA H100 SXM 80GB: VRAM Requirement & Speed Estimate
Yes. At Q4_K_M 4-bit quantization with a 16K context window, Qwen3-32B requires 25.03 GB of VRAM (19.88 GB weights + 4.29 GB KV cache + 0.85 GB overhead), fitting within H100 80G's 78.4 GB usable memory at ~109.5 tokens/s.
Concurrency (Batch):
Dense 32.8B·64L (d=5120)·64Q/8KV·Ctx 128K
Smooth Fit25.03 / 78.4 GB~109.5 tok/s· Q4_K_M · 16K ctx
3. Weight Quantization
4. Per-Request Context16K (16,384 tok)
Verified: 2026-11-20·Roofline bandwidth/compute estimate (Batch=1)
Smooth Fit
25.03/ 78.4 GB usable
Physical 80 GB · Load 31.9%
Theoretical Decode Speed
~109.5tokens/s
Actual speed varies ~±25% by runtime
Max Context at Batch=1
128K/ 128K
Enable Q8/Q4 KV quantization to scale batch concurrency
Weights: 19.88 GBKV Cache (1×16K): 4.29 GBOverhead: 0.85 GB
🚀 Blazing Fast (Instant CoT)~4.6s for a 500-token response