🔌
New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
// RUN ON DATACENTER GPU
Run Llama 3.2 90B on Datacenter GPU hardware
10 options ranked by estimated speed at Q4 / 8k context.
| Hardware | Est tok/s | Fits 8k? | Max ctx | ~Price |
|---|---|---|---|---|
| NVIDIA B200 → | ~140.3 | yes | 383k | — |
| AMD Instinct MI325X → | ~93.1 | yes | 563k | — |
| NVIDIA H200 → | ~84.2 | yes | 240k | — |
| AMD Instinct MI300X 192GB → | ~82.3 | yes | 383k | ~$15,000 |
| NVIDIA H100 SXM5 80GB → | ~58.7 | yes | 69k | ~$30,000 |
| Intel Gaudi 3 → | ~53.7 | yes | 203k | — |
| NVIDIA A100 SXM4 80GB → | ~35.8 | yes | 69k | ~$15,000 |
| NVIDIA H100 PCIe 80GB → | ~35.1 | yes | 69k | ~$25,000 |
| NVIDIA A100 PCIe 80GB → | ~33.9 | yes | 69k | ~$12,000 |
| NVIDIA RTX PRO 6000 Blackwell → | ~31.4 | yes | 114k | ~$8,500 |
Estimate at Q4 (memory-bound). Click any option for its full per-hardware estimate and measured numbers.
FAQ
What Datacenter GPU hardware runs Llama 3.2 90B?▶
On GenAIList, 10 Datacenter GPU options can run Llama 3.2 90B at Q4 — the fastest is the NVIDIA B200 at ~140.3 tok/s.
How fast is Llama 3.2 90B on Datacenter GPU hardware?▶
Estimated generation speed ranges from ~31.4 to ~140.3 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.