New: connect Claude & other AIs to GenAIList over MCP โ research the catalog and contribute to the shared knowledge base. Learn how โ
Run Evo 2 40B 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 โ | ~308.4 | yes | 466k | โ |
| AMD Instinct MI325X โ | ~204.7 | yes | 645k | โ |
| NVIDIA H200 โ | ~185 | yes | 322k | โ |
| AMD Instinct MI300X 192GB โ | ~180.8 | yes | 466k | ~$15,000 |
| NVIDIA H100 SXM5 80GB โ | ~129.1 | yes | 151k | ~$30,000 |
| Intel Gaudi 3 โ | ~118 | yes | 286k | โ |
| NVIDIA A100 SXM4 80GB โ | ~78.6 | yes | 151k | ~$15,000 |
| NVIDIA H100 PCIe 80GB โ | ~77.1 | yes | 151k | ~$25,000 |
| NVIDIA A100 PCIe 80GB โ | ~74.6 | yes | 151k | ~$12,000 |
| NVIDIA RTX PRO 6000 Blackwell โ | ~69.1 | yes | 196k | ~$8,500 |
| NVIDIA RTX 6000 Ada 48GB โ | ~37 | yes | 61k | ~$6,800 |
| NVIDIA L40S 48GB โ | ~33.3 | yes | 61k | ~$10,000 |
| AMD Radeon Pro W7900 48GB โ | ~29.5 | yes | 61k | ~$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 Evo 2 40B?โถ
On GenAIList, 13 Datacenter GPU options can run Evo 2 40B at Q4 โ the fastest is the NVIDIA B200 at ~308.4 tok/s.
How fast is Evo 2 40B on Datacenter GPU hardware?โถ
Estimated generation speed ranges from ~29.5 to ~308.4 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.