
NVIDIA A100 80GB PCIe (New)
Still the best price-to-VRAM ratio in the new-unit datacenter segment. A100 remains the standard for fine-tuning, MIG-partitioned multi-tenant serving, and CUDA workloads that do not need Hopper features.
MOQ 1 unit
| Architecture | NVIDIA Ampere (GA100) |
| CUDA cores | 6,912 |
| Tensor cores | 432 (3rd gen) |
| Memory | 80 GB HBM2e |
| Memory bandwidth | 1.94 TB/s |
| FP16 performance | 624 TFLOPS (with sparsity) |
| TF32 performance | 312 TFLOPS (with sparsity) |
| FP64 performance | 19.5 TFLOPS (Tensor) |
| Interconnect | NVLink 600 GB/s + PCIe 4.0 x16 |
| MIG support | Up to 7 instances |
| TDP | 300 W |
| Form factor | Dual-slot FHFL, passive cooling |
| Warranty | 12 months, SYNDROME AG |
Still the best price-to-VRAM ratio in the new-unit datacenter segment. A100 remains the standard for fine-tuning, MIG-partitioned multi-tenant serving, and CUDA workloads that do not need Hopper features.
32 units are on the shelf across Singapore and Dubai as of the last price update. Pricing is FOB Singapore / Dubai in USD; volume orders from 10 units receive −5…10%. Serial numbers are recorded for every unit before it leaves the warehouse.
Air and sea freight from Singapore or Dubai under EXW, FOB, CIF or DDP (Incoterms 2020). Export clearance and shipping documents are handled by our hubs. Commercial terms are structured individually for each deal and jurisdiction — your manager will walk you through the options.
New hardware carries a 12-month warranty (refurbished — 6 months). RMA is handled through the nearest hub with advance replacement from stock where available. Serial numbers are recorded for every unit that leaves the warehouse.
