🔌
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 mBART-50 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 → | ~20,374.7 | yes | 2667k | — |
| AMD Instinct MI325X → | ~13,524.6 | yes | 3566k | — |
| NVIDIA H200 → | ~12,224.8 | yes | 1951k | — |
| AMD Instinct MI300X 192GB → | ~11,946.7 | yes | 2667k | ~$15,000 |
| NVIDIA H100 SXM5 80GB → | ~8,531.9 | yes | 1095k | ~$30,000 |
| Intel Gaudi 3 → | ~7,798.6 | yes | 1769k | — |
| NVIDIA A100 SXM4 80GB → | ~5,193 | yes | 1095k | ~$15,000 |
| NVIDIA H100 PCIe 80GB → | ~5,093.7 | yes | 1095k | ~$25,000 |
| NVIDIA A100 PCIe 80GB → | ~4,928.1 | yes | 1095k | ~$12,000 |
| NVIDIA RTX PRO 6000 Blackwell → | ~4,563.9 | yes | 1320k | ~$8,500 |
| NVIDIA RTX 6000 Ada 48GB → | ~2,445 | yes | 646k | ~$6,800 |
| NVIDIA L40S 48GB → | ~2,200.5 | yes | 646k | ~$10,000 |
| AMD Radeon Pro W7900 48GB → | ~1,947.5 | yes | 646k | ~$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 mBART-50?▶
On GenAIList, 13 Datacenter GPU options can run mBART-50 at Q4 — the fastest is the NVIDIA B200 at ~20,374.7 tok/s.
How fast is mBART-50 on Datacenter GPU hardware?▶
Estimated generation speed ranges from ~1,947.5 to ~20,374.7 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.