🔌
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 Qwen2.5-72B on Datacenter GPU hardware
13 options ranked by estimated speed at Q4 / 8k context.
| Hardware | Est tok/s | Fits 8k? | Max ctx | ~Price |
|---|---|---|---|---|
| NVIDIA B200 → | ~171 | yes | 410k | — |
| AMD Instinct MI325X → | ~113.5 | yes | 590k | — |
| NVIDIA H200 → | ~102.6 | yes | 267k | — |
| AMD Instinct MI300X 192GB → | ~100.2 | yes | 410k | ~$15,000 |
| NVIDIA H100 SXM5 80GB → | ~71.6 | yes | 96k | ~$30,000 |
| Intel Gaudi 3 → | ~65.4 | yes | 231k | — |
| NVIDIA A100 SXM4 80GB → | ~43.6 | yes | 96k | ~$15,000 |
| NVIDIA H100 PCIe 80GB → | ~42.7 | yes | 96k | ~$25,000 |
| NVIDIA A100 PCIe 80GB → | ~41.4 | yes | 96k | ~$12,000 |
| NVIDIA RTX PRO 6000 Blackwell → | ~38.3 | yes | 141k | ~$8,500 |
| NVIDIA RTX 6000 Ada 48GB → | ~20.5 | tight | 6k | ~$6,800 |
| NVIDIA L40S 48GB → | ~18.5 | tight | 6k | ~$10,000 |
| AMD Radeon Pro W7900 48GB → | ~16.3 | tight | 6k | ~$3,999 |
Estimate at Q4 (memory-bound). Click any option for its full per-hardware estimate and measured numbers.
FAQ
What Datacenter GPU hardware runs Qwen2.5-72B?▶
On GenAIList, 13 Datacenter GPU options can run Qwen2.5-72B at Q4 — the fastest is the NVIDIA B200 at ~171 tok/s.
How fast is Qwen2.5-72B on Datacenter GPU hardware?▶
Estimated generation speed ranges from ~16.3 to ~171 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.