🔌
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 RWKV-4 World (7B) 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 → | ~1,681.1 | yes | 1305k | — |
| AMD Instinct MI325X → | ~1,115.9 | yes | 1754k | — |
| NVIDIA H200 → | ~1,008.7 | yes | 947k | — |
| AMD Instinct MI300X 192GB → | ~985.7 | yes | 1305k | ~$15,000 |
| NVIDIA H100 SXM5 80GB → | ~704 | yes | 518k | ~$30,000 |
| Intel Gaudi 3 → | ~643.5 | yes | 855k | — |
| NVIDIA A100 SXM4 80GB → | ~428.5 | yes | 518k | ~$15,000 |
| NVIDIA H100 PCIe 80GB → | ~420.3 | yes | 518k | ~$25,000 |
| NVIDIA A100 PCIe 80GB → | ~406.6 | yes | 518k | ~$12,000 |
| NVIDIA RTX PRO 6000 Blackwell → | ~376.6 | yes | 631k | ~$8,500 |
| NVIDIA RTX 6000 Ada 48GB → | ~201.7 | yes | 294k | ~$6,800 |
| NVIDIA L40S 48GB → | ~181.6 | yes | 294k | ~$10,000 |
| AMD Radeon Pro W7900 48GB → | ~160.7 | yes | 294k | ~$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 RWKV-4 World (7B)?▶
On GenAIList, 13 Datacenter GPU options can run RWKV-4 World (7B) at Q4 — the fastest is the NVIDIA B200 at ~1,681.1 tok/s.
How fast is RWKV-4 World (7B) on Datacenter GPU hardware?▶
Estimated generation speed ranges from ~160.7 to ~1,681.1 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.