| 规格 | NVIDIA T4 | NVIDIA A30 |
|---|
| 架构 | Turing (TU104-895) | NVIDIA Ampere |
| 显存 | 16GB GDDR6 with ECC (默认开启) | 24 GB HBM2 |
| 显存带宽 | 320GB/s (早期一页版datasheet口径300GB/s) | 933 GB/s |
| 显存位宽 | 256bit | — |
| 算力 | FP32 8.1 TFLOPS / FP16·FP32混合精度 65 TFLOPS / INT8 130 TOPS / INT4 260 TOPS | — |
| CUDA核心 | 2,560 (40 SM) / Tensor Core 320 / RT Core 40 (白皮书口径) | — |
| 功耗 | 70W (Total board power, 默认) | — |
| 互联 | PCIe Gen3 x16 (支持x8运行, Interconnect BW 32GB/s) | PCIe Gen4: 64GB/s / Third-gen NVLINK: 200GB/s |
| 发布年 | 2018 (Product Brief v01 2018-10-24初版, Dec18版datasheet © 2018 NVIDIA) | — |
| 形态 | 单槽半高半长 Low-Profile 6.6英寸(长169.53mm), 被动散热, 板重301g | Dual-slot, full-height, full-length (FHFL) |
| 时钟 | Base 585MHz / Boost 1590MHz (Base为70W功耗墙下调校) | — |
| 媒体引擎 | 1080p30解码 H.264 32路 / HEVC 44路 / VP9 32路 (白皮书Table 2, 官方页称38路full-HD) | — |
| SR-IOV | Supported; 16 VF（支持） | — |
| 工艺 | 12nm FFN, 136亿晶体管, 545mm2 (白皮书TU104口径) | — |
| Compute APIs | CUDA, NVIDIA TensorRT, ONNX | — |
| FP32算力 | — | 10.3 teraFLOPS |
| FP64算力 | — | 5.2 teraFLOPS |
| TF32算力 | — | 82 teraFLOPS | 165 teraFLOPS |
| FP16算力 | — | 165 teraFLOPS | 330 teraFLOPS |
| INT8算力 | — | 330 TOPS | 661 TOPS |
| 功耗TDP | — | 165 W |
| MIG | — | 4 GPU instances @ 6GB each |