🔌
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 iGPT-L 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 → | ~9,125.2 | yes | 2661k | — |
| AMD Instinct MI325X → | ~6,057.3 | yes | 3559k | — |
| NVIDIA H200 → | ~5,475.1 | yes | 1945k | — |
| AMD Instinct MI300X 192GB → | ~5,350.6 | yes | 2661k | ~$15,000 |
| NVIDIA H100 SXM5 80GB → | ~3,821.2 | yes | 1089k | ~$30,000 |
| Intel Gaudi 3 → | ~3,492.8 | yes | 1762k | — |
| NVIDIA A100 SXM4 80GB → | ~2,325.8 | yes | 1089k | ~$15,000 |
| NVIDIA H100 PCIe 80GB → | ~2,281.3 | yes | 1089k | ~$25,000 |
| NVIDIA A100 PCIe 80GB → | ~2,207.2 | yes | 1089k | ~$12,000 |
| NVIDIA RTX PRO 6000 Blackwell → | ~2,044.1 | yes | 1313k | ~$8,500 |
| NVIDIA RTX 6000 Ada 48GB → | ~1,095 | yes | 639k | ~$6,800 |
| NVIDIA L40S 48GB → | ~985.5 | yes | 639k | ~$10,000 |
| AMD Radeon Pro W7900 48GB → | ~872.2 | yes | 639k | ~$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 iGPT-L?▶
On GenAIList, 13 Datacenter GPU options can run iGPT-L at Q4 — the fastest is the NVIDIA B200 at ~9,125.2 tok/s.
How fast is iGPT-L on Datacenter GPU hardware?▶
Estimated generation speed ranges from ~872.2 to ~9,125.2 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.