🔌
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 ERNIE-4.5-VL-28B-A3B 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 → | ~443.9 | yes | 608k | — |
| AMD Instinct MI325X → | ~294.6 | yes | 833k | — |
| NVIDIA H200 → | ~266.3 | yes | 429k | — |
| AMD Instinct MI300X 192GB → | ~260.3 | yes | 608k | ~$15,000 |
| NVIDIA H100 SXM5 80GB → | ~185.9 | yes | 215k | ~$30,000 |
| Intel Gaudi 3 → | ~169.9 | yes | 384k | — |
| NVIDIA A100 SXM4 80GB → | ~113.1 | yes | 215k | ~$15,000 |
| NVIDIA H100 PCIe 80GB → | ~111 | yes | 215k | ~$25,000 |
| NVIDIA A100 PCIe 80GB → | ~107.4 | yes | 215k | ~$12,000 |
| NVIDIA RTX PRO 6000 Blackwell → | ~99.4 | yes | 271k | ~$8,500 |
| NVIDIA RTX 6000 Ada 48GB → | ~53.3 | yes | 103k | ~$6,800 |
| NVIDIA L40S 48GB → | ~47.9 | yes | 103k | ~$10,000 |
| AMD Radeon Pro W7900 48GB → | ~42.4 | yes | 103k | ~$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 ERNIE-4.5-VL-28B-A3B?▶
On GenAIList, 13 Datacenter GPU options can run ERNIE-4.5-VL-28B-A3B at Q4 — the fastest is the NVIDIA B200 at ~443.9 tok/s.
How fast is ERNIE-4.5-VL-28B-A3B on Datacenter GPU hardware?▶
Estimated generation speed ranges from ~42.4 to ~443.9 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.