| 规格 | NVIDIA RTX A5000 | NVIDIA A30 |
|---|
| 架构 | Ampere (GA102, GA102-850-A1) | NVIDIA Ampere |
| 显存 | 24GB GDDR6 with ECC | 24 GB HBM2 |
| 显存带宽 | 768GB/s | 933 GB/s |
| 显存位宽 | 384bit | — |
| 算力 | FP32 27.8 TFLOPS / RT Core 54.2 TFLOPS / Tensor 222.2 TFLOPS (含稀疏加速) | — |
| CUDA核心 | 8192 / Tensor Core 256 (三代) / RT Core 64 (二代) | — |
| 功耗 | 230W (主动散热, 1x 8-pin PCIe供电) | — |
| 互联 | NVLink 112.5GB/s 双向 (低矮桥接器2槽/3槽, 双卡拼48GB, 桥接器另购) / PCIe Gen4 x16 | PCIe Gen4: 64GB/s / Third-gen NVLINK: 200GB/s |
| 发布年 | 2021 (2021-04-12, 与RTX A4000同日发布) | — |
| 形态 | Dual-slot 双槽全高, 4.4"H x 10.5"L, Active主动散热 | Dual-slot, full-height, full-length (FHFL) |
| 显示输出 | 4x DisplayPort 1.4a | — |
| 编解码 | 1 Encode / 2 Decode (+AV1解码) | — |
| VR | Yes (VR Ready, vGPU 软件支持 vPC/vApps/RTX vWS/vCS) | — |
| 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 |