🔌
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 BART-large 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 → | ~30,590.3 | yes | 2669k | — |
| AMD Instinct MI325X → | ~20,305.6 | yes | 3567k | — |
| NVIDIA H200 → | ~18,354.2 | yes | 1953k | — |
| AMD Instinct MI300X 192GB → | ~17,936.6 | yes | 2669k | ~$15,000 |
| NVIDIA H100 SXM5 80GB → | ~12,809.7 | yes | 1097k | ~$30,000 |
| Intel Gaudi 3 → | ~11,708.7 | yes | 1771k | — |
| NVIDIA A100 SXM4 80GB → | ~7,796.7 | yes | 1097k | ~$15,000 |
| NVIDIA H100 PCIe 80GB → | ~7,647.6 | yes | 1097k | ~$25,000 |
| NVIDIA A100 PCIe 80GB → | ~7,399 | yes | 1097k | ~$12,000 |
| NVIDIA RTX PRO 6000 Blackwell → | ~6,852.2 | yes | 1321k | ~$8,500 |
| NVIDIA RTX 6000 Ada 48GB → | ~3,670.8 | yes | 647k | ~$6,800 |
| NVIDIA L40S 48GB → | ~3,303.8 | yes | 647k | ~$10,000 |
| AMD Radeon Pro W7900 48GB → | ~2,924 | yes | 647k | ~$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 BART-large?▶
On GenAIList, 13 Datacenter GPU options can run BART-large at Q4 — the fastest is the NVIDIA B200 at ~30,590.3 tok/s.
How fast is BART-large on Datacenter GPU hardware?▶
Estimated generation speed ranges from ~2,924 to ~30,590.3 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.