RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can Qwen3 32B run on the A100 80GB?

Yes. At 8K context the best fit is F16, using about 63.7 GB of the 76.0 GB usable on the A100 80GB. Estimated speed ceiling: ~31 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctxMax tok/s (est., 8K)
2-bit (Q2_K)13.1 GB Fits14.1 GB Fits20.1 GB Fits166
3-bit (Q3_K_M)16.6 GB Fits17.6 GB Fits23.6 GB Fits127
4-bit (Q4_K_M) •20.1 GB Fits21.1 GB Fits27.1 GB Fits103
5-bit (Q5_K_M) •23.3 GB Fits24.3 GB Fits30.3 GB Fits88
6-bit (Q6_K) •26.7 GB Fits27.7 GB Fits33.7 GB Fits76
8-bit (Q8_0) •34.1 GB Fits35.1 GB Fits41.1 GB Fits59
16-bit (F16/BF16)62.7 GB Fits63.7 GB Fits69.7 GB Tight31

• real GGUF file size from Qwen/Qwen3-32B-GGUF. Speed is a bandwidth ceiling (2039 GB/s), not a benchmark. Methodology.

More

Qwen3 32B on every GPU and Mac · Every model that runs on the A100 80GB · Other models that fit 64 GB VRAM · Try other settings in the calculator

Data: Hugging Face Hub (Qwen/Qwen3-32B), NVIDIA specs. Data as of 2026-11-17.