🔌
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 TeleChat2-35B 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 → | ~355.1 | yes | 593k | — |
| AMD Instinct MI325X → | ~235.7 | yes | 818k | — |
| NVIDIA H200 → | ~213.1 | yes | 414k | — |
| AMD Instinct MI300X 192GB → | ~208.2 | yes | 593k | ~$15,000 |
| NVIDIA H100 SXM5 80GB → | ~148.7 | yes | 200k | ~$30,000 |
| Intel Gaudi 3 → | ~135.9 | yes | 369k | — |
| NVIDIA A100 SXM4 80GB → | ~90.5 | yes | 200k | ~$15,000 |
| NVIDIA H100 PCIe 80GB → | ~88.8 | yes | 200k | ~$25,000 |
| NVIDIA A100 PCIe 80GB → | ~85.9 | yes | 200k | ~$12,000 |
| NVIDIA RTX PRO 6000 Blackwell → | ~79.5 | yes | 256k | ~$8,500 |
| NVIDIA RTX 6000 Ada 48GB → | ~42.6 | yes | 88k | ~$6,800 |
| NVIDIA L40S 48GB → | ~38.4 | yes | 88k | ~$10,000 |
| AMD Radeon Pro W7900 48GB → | ~33.9 | yes | 88k | ~$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 TeleChat2-35B?▶
On GenAIList, 13 Datacenter GPU options can run TeleChat2-35B at Q4 — the fastest is the NVIDIA B200 at ~355.1 tok/s.
How fast is TeleChat2-35B on Datacenter GPU hardware?▶
Estimated generation speed ranges from ~33.9 to ~355.1 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.