🔌
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 base LM+GNN+kNN 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 → | ~45,359.7 | yes | 2670k | — |
| AMD Instinct MI325X → | ~30,109.5 | yes | 3569k | — |
| NVIDIA H200 → | ~27,215.8 | yes | 1954k | — |
| AMD Instinct MI300X 192GB → | ~26,596.7 | yes | 2670k | ~$15,000 |
| NVIDIA H100 SXM5 80GB → | ~18,994.4 | yes | 1098k | ~$30,000 |
| Intel Gaudi 3 → | ~17,361.8 | yes | 1772k | — |
| NVIDIA A100 SXM4 80GB → | ~11,561.1 | yes | 1098k | ~$15,000 |
| NVIDIA H100 PCIe 80GB → | ~11,339.9 | yes | 1098k | ~$25,000 |
| NVIDIA A100 PCIe 80GB → | ~10,971.4 | yes | 1098k | ~$12,000 |
| NVIDIA RTX PRO 6000 Blackwell → | ~10,160.6 | yes | 1322k | ~$8,500 |
| NVIDIA RTX 6000 Ada 48GB → | ~5,443.2 | yes | 649k | ~$6,800 |
| NVIDIA L40S 48GB → | ~4,898.9 | yes | 649k | ~$10,000 |
| AMD Radeon Pro W7900 48GB → | ~4,335.8 | yes | 649k | ~$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 base LM+GNN+kNN?▶
On GenAIList, 13 Datacenter GPU options can run base LM+GNN+kNN at Q4 — the fastest is the NVIDIA B200 at ~45,359.7 tok/s.
How fast is base LM+GNN+kNN on Datacenter GPU hardware?▶
Estimated generation speed ranges from ~4,335.8 to ~45,359.7 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.