🔌
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 Qwen2.5-14B 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 → | ~845.5 | yes | 1019k | — |
| AMD Instinct MI325X → | ~561.2 | yes | 1378k | — |
| NVIDIA H200 → | ~507.3 | yes | 732k | — |
| AMD Instinct MI300X 192GB → | ~495.7 | yes | 1019k | ~$15,000 |
| NVIDIA H100 SXM5 80GB → | ~354 | yes | 390k | ~$30,000 |
| Intel Gaudi 3 → | ~323.6 | yes | 659k | — |
| NVIDIA A100 SXM4 80GB → | ~215.5 | yes | 390k | ~$15,000 |
| NVIDIA H100 PCIe 80GB → | ~211.4 | yes | 390k | ~$25,000 |
| NVIDIA A100 PCIe 80GB → | ~204.5 | yes | 390k | ~$12,000 |
| NVIDIA RTX PRO 6000 Blackwell → | ~189.4 | yes | 480k | ~$8,500 |
| NVIDIA RTX 6000 Ada 48GB → | ~101.5 | yes | 210k | ~$6,800 |
| NVIDIA L40S 48GB → | ~91.3 | yes | 210k | ~$10,000 |
| AMD Radeon Pro W7900 48GB → | ~80.8 | yes | 210k | ~$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 Qwen2.5-14B?▶
On GenAIList, 13 Datacenter GPU options can run Qwen2.5-14B at Q4 — the fastest is the NVIDIA B200 at ~845.5 tok/s.
How fast is Qwen2.5-14B on Datacenter GPU hardware?▶
Estimated generation speed ranges from ~80.8 to ~845.5 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.