| 规格 | NVIDIA A30 | NVIDIA H20 |
|---|
| 架构 | NVIDIA Ampere | NVIDIA Hopper |
| 显存 | 24 GB HBM2 | 96GB HBM3(Flopper/VipraTech/腾讯新闻三源一致) |
| 显存带宽 | 933 GB/s | 4.0TB/s |
| 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 | — |
| 互联 | PCIe Gen4: 64GB/s / Third-gen NVLINK: 200GB/s | NVLink 900GB/s+PCIe Gen5 x16 128GB/s |
| 形态 | Dual-slot, full-height, full-length (FHFL) | HGX 8路SXM基板模组(Flopper'SXM'/VipraTech'8-way HGX'/腾讯新闻'HGX方案8路GPU配置+H20 SXM';同批L20/L2为PCIe形态) |
| MIG | 4 GPU instances @ 6GB each | — |
| 算力 | — | FP8 Tensor Core 296 TFLOPS(密集)/592 TFLOPS(2:4稀疏,单源);FP16/BF16 Tensor Core 148 TFLOPS;TF32 Tensor Core 74 TFLOPS;FP32 44 TFLOPS;FP64 1.0 TFLOPS(合规裁剪:H100 SXM同表FP8 3958/FP16 1979/FP64 34);INT8 296 TOPS/592 TOPS(单源) |
| 功耗 | — | 400W TDP(Flopper/VipraTech规格表/腾讯新闻'热设计功耗400W'三源;VipraTech正文另有350W矛盾口径已弃) |
| 发布年 | — | 2024(Flopper核验Launch Year 2024;腾讯新闻2023-11-10首发报道'即将推出';具体发布日未核验双源,不录) |
| L2缓存 | — | 60MB(单源VipraTech) |
| CUDA核心 | — | 14,592 FP32 CUDA核心(单源VipraTech) |
| MIG切片 | — | 最多7实例(VipraTech+腾讯新闻'7路MIG'双源) |
| 编解码器 | — | 7 NVDEC + 7 NVJPEG(单源VipraTech) |
| 出口合规背景 | — | 2022-10美国出口管制后对华合规特供卡:算力密度较H100大幅裁剪(FP8约为H100 SXM的7.5%),以显存96GB+NVLink900GB/s保推理场景(QQ:综合算力约H100的20%) |