🔌
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 TinyBert 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 → | ~185,501.1 | yes | 2672k | — |
| AMD Instinct MI325X → | ~123,134.3 | yes | 3570k | — |
| NVIDIA H200 → | ~111,300.6 | yes | 1956k | — |
| AMD Instinct MI300X 192GB → | ~108,768.7 | yes | 2672k | ~$15,000 |
| NVIDIA H100 SXM5 80GB → | ~77,678.6 | yes | 1100k | ~$30,000 |
| Intel Gaudi 3 → | ~71,002.1 | yes | 1773k | — |
| NVIDIA A100 SXM4 80GB → | ~47,279.6 | yes | 1100k | ~$15,000 |
| NVIDIA H100 PCIe 80GB → | ~46,375.3 | yes | 1100k | ~$25,000 |
| NVIDIA A100 PCIe 80GB → | ~44,868.1 | yes | 1100k | ~$12,000 |
| NVIDIA RTX PRO 6000 Blackwell → | ~41,552.2 | yes | 1324k | ~$8,500 |
| NVIDIA RTX 6000 Ada 48GB → | ~22,260.1 | yes | 650k | ~$6,800 |
| NVIDIA L40S 48GB → | ~20,034.1 | yes | 650k | ~$10,000 |
| AMD Radeon Pro W7900 48GB → | ~17,731.3 | yes | 650k | ~$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 TinyBert?▶
On GenAIList, 13 Datacenter GPU options can run TinyBert at Q4 — the fastest is the NVIDIA B200 at ~185,501.1 tok/s.
How fast is TinyBert on Datacenter GPU hardware?▶
Estimated generation speed ranges from ~17,731.3 to ~185,501.1 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.