XPU Catalog
Browse all AI compute accelerators with verified specs and benchmarks
Found 128 XPUs
| Vendor↕ | Model↕ | Architecture↕ | Form Factor↕ | VRAM (GB)↑ | TFLOPs↕ | TDP (W)↕ | Launch Date↕ | Actions |
|---|---|---|---|---|---|---|---|---|
| NVIDIA | GeForce GT 710 | Kepler | PCIe | 2 | 0.366 | 19 | Mar 2014 | Details |
| NVIDIA | GeForce GT 730 | Kepler | PCIe | 2 | 0.693 | 49 | Jun 2014 | Details |
| NVIDIA | Quadro K620 | Kepler | PCIe | 2 | 0.768 | 45 | Jul 2014 | Details |
| NVIDIA | Quadro P600 | Pascal | PCIe | 2 | 1.117 | 40 | Feb 2017 | Details |
| NVIDIA | Quadro P620 | Pascal | PCIe | 2 | 1.386 | 40 | Feb 2017 | Details |
| Graphcore | Bow IPU | Bow-2000 | — | 3.6 | — | 200 | Mar 2022 | Details |
| Graphcore | IPU-M2000 | Colossus Mk2 | — | 3.6 | — | 180 | Jul 2020 | Details |
| NVIDIA | GeForce GTX 1050 Ti | Pascal | PCIe | 4 | 2.138 | 75 | Oct 2016 | Details |
| NVIDIA | GeForce GTX 1650 | Turing | PCIe | 4 | 2.984 | 75 | Apr 2019 | Details |
| NVIDIA | Quadro P1000 | Pascal | PCIe | 4 | 1.894 | 47 | Feb 2017 | Details |
| NVIDIA | GeForce GTX 1060 | Pascal | PCIe | 6 | 4.375 | 120 | Jul 2016 | Details |
| NVIDIA | GeForce GTX 1660 | Turing | PCIe | 6 | 5.027 | 120 | Mar 2019 | Details |
| NVIDIA | GeForce GTX 1660 Super | Turing | PCIe | 6 | 5.027 | 125 | Oct 2019 | Details |
| NVIDIA | GeForce RTX 2060 | Turing | PCIe | 6 | 6.5 | 160 | Jan 2019 | Details |
| NVIDIA | GeForce GTX 1070 | Pascal | PCIe | 8 | 6.463 | 150 | Jun 2016 | Details |
| NVIDIA | GeForce GTX 1070 Ti | Pascal | PCIe | 8 | 8.186 | 180 | Nov 2017 | Details |
| NVIDIA | GeForce GTX 1080 | Pascal | PCIe | 8 | 8.873 | 180 | May 2016 | Details |
| NVIDIA | GeForce RTX 2060 Super | Turing | PCIe | 8 | 7.2 | 175 | Jul 2019 | Details |
| NVIDIA | GeForce RTX 2070 | Turing | PCIe | 8 | 7.5 | 175 | Oct 2018 | Details |
| NVIDIA | GeForce RTX 2080 | Turing | PCIe | 8 | 10.1 | 215 | Sep 2018 | Details |
| NVIDIA | GeForce RTX 3050 | Ampere | PCIe | 8 | 18.2 | 130 | Jan 2022 | Details |
| NVIDIA | GeForce RTX 3060 Ti | Ampere | PCIe | 8 | 33.2 | 200 | Dec 2020 | Details |
| NVIDIA | GeForce RTX 3070 | Ampere | PCIe | 8 | 40.5 | 220 | Oct 2020 | Details |
| NVIDIA | GeForce RTX 3070 Ti | Ampere | PCIe | 8 | 43.2 | 290 | Jun 2021 | Details |
| NVIDIA | GeForce RTX 4060 | Ada Lovelace | PCIe | 8 | 30 | 115 | Jun 2023 | Details |
| NVIDIA | GeForce RTX 4060 Ti | Ada Lovelace | PCIe | 8 | 44 | 160 | May 2023 | Details |
| NVIDIA | GeForce RTX 5060 | Blackwell | PCIe | 8 | 60 | 170 | Mar 2025 | Details |
| Tenstorrent | Grayskull | Tensix Core | — | 8 | 200 | 65 | Mar 2021 | Details |
| NVIDIA | P4 | Pascal | PCIe | 8 | 5.5 | 75 | Sep 2016 | Details |
| NVIDIA | Quadro M4000 | Maxwell | PCIe | 8 | 2.57 | 120 | Jun 2015 | Details |
| NVIDIA | Quadro P4000 | Pascal | PCIe | 8 | 5.304 | 105 | Oct 2016 | Details |
| NVIDIA | RTX 4000 | Turing | PCIe | 8 | 57.6 | 160 | Nov 2018 | Details |
| NVIDIA | T1000 | Turing | PCIe | 8 | 2.5 | 50 | May 2019 | Details |
| NVIDIA | GeForce RTX 3080 | Ampere | PCIe | 10 | 59.2 | 320 | Sep 2020 | Details |
| NVIDIA | GeForce GTX 1080 Ti | Pascal | PCIe | 11 | 11.34 | 250 | Mar 2017 | Details |
| NVIDIA | GeForce RTX 2080 Ti | Turing | PCIe | 11 | 13.4 | 250 | Sep 2018 | Details |
| NVIDIA | GeForce RTX 3060 | Ampere | PCIe | 12 | 25.4 | 170 | Feb 2021 | Details |
| NVIDIA | GeForce RTX 3080 Ti | Ampere | PCIe | 12 | 68.4 | 350 | Jun 2021 | Details |
| NVIDIA | GeForce RTX 4070 | Ada Lovelace | PCIe | 12 | 58.92 | 200 | Apr 2023 | Details |
| NVIDIA | GeForce RTX 4070 Super | Ada Lovelace | PCIe | 12 | 70 | 220 | Jan 2024 | Details |
| NVIDIA | GeForce RTX 4070 Ti | Ada Lovelace | PCIe | 12 | 80.16 | 285 | Jan 2023 | Details |
| NVIDIA | GeForce RTX 5070 | Blackwell | PCIe | 12 | 110 | 250 | Feb 2025 | Details |
| NVIDIA | RTX A2000 | Ampere | PCIe | 12 | 16 | 70 | Oct 2021 | Details |
| NVIDIA | GeForce RTX 4060 Ti 16GB | Ada Lovelace | PCIe | 16 | 44 | 165 | Jul 2024 | Details |
| NVIDIA | GeForce RTX 4070 Ti Super | Ada Lovelace | PCIe | 16 | 88 | 285 | Jan 2024 | Details |
| NVIDIA | GeForce RTX 4080 | Ada Lovelace | PCIe | 16 | 97.76 | 320 | Nov 2022 | Details |
| NVIDIA | GeForce RTX 4080 Super | Ada Lovelace | PCIe | 16 | 104 | 320 | Jan 2024 | Details |
| NVIDIA | GeForce RTX 5060 Ti | Blackwell | PCIe | 16 | 88 | 220 | Mar 2025 | Details |
| NVIDIA | GeForce RTX 5070 Ti | Blackwell | PCIe | 16 | 133 | 300 | Feb 2025 | Details |
| NVIDIA | GeForce RTX 5080 | Blackwell | PCIe | 16 | 171 | 360 | Jan 2025 | Details |
| Alibaba | Hanguang 800 | XuanTie | — | 16 | — | 150 | Sep 2019 | Details |
| NVIDIA | P100 | Pascal | SXM | 16 | 9.3 | 300 | Apr 2016 | Details |
| NVIDIA | Quadro M60 | Maxwell | PCIe | 16 | 4.825 | 300 | Mar 2016 | Details |
| NVIDIA | Quadro P5000 | Pascal | PCIe | 16 | 8.873 | 180 | Oct 2016 | Details |
| NVIDIA | RTX 5000 | Turing | PCIe | 16 | 89.2 | 265 | Aug 2018 | Details |
| NVIDIA | RTX A4000 | Ampere | PCIe | 16 | 38.7 | 140 | Apr 2021 | Details |
| NVIDIA | T4 | Turing | PCIe | 16 | 65 | 70 | Sep 2018 | Details |
| NVIDIA | T4G | Turing | PCIe | 16 | 65 | 70 | May 2020 | Details |
| TPU v5e | TPU v5 | Mezzanine | 16 | 197 | 250 | Aug 2023 | Details | |
| NVIDIA | RTX 4000 SFF Ada Generation | Ada Lovelace | PCIe | 20 | 52 | 70 | Apr 2023 | Details |
| NVIDIA | RTX A4500 | Ampere | PCIe | 20 | 46.6 | 200 | Oct 2021 | Details |
| NVIDIA | A10 | Ampere | PCIe | 24 | 125 | 150 | Apr 2021 | Details |
| NVIDIA | A10G | Ampere | PCIe | 24 | 125 | 300 | Jul 2021 | Details |
| NVIDIA | A30 | Ampere | PCIe | 24 | 165 | 165 | Apr 2021 | Details |
| NVIDIA | GeForce RTX 3090 | Ampere | PCIe | 24 | 71 | 350 | Sep 2020 | Details |
| NVIDIA | GeForce RTX 3090 Ti | Ampere | PCIe | 24 | 80 | 450 | Mar 2022 | Details |
| NVIDIA | GeForce RTX 4090 | Ada Lovelace | PCIe | 24 | 165.2 | 450 | Oct 2022 | Details |
| NVIDIA | GeForce RTX 4090 D | Ada Lovelace | PCIe | 24 | 165 | 425 | Jan 2024 | Details |
| NVIDIA | K80 | Kepler | PCIe | 24 | 8.73 | 300 | Nov 2014 | Details |
| NVIDIA | L4 | Ada Lovelace | PCIe | 24 | 121 | 72 | Mar 2023 | Details |
| NVIDIA | P40 | Pascal | PCIe | 24 | 12 | 250 | Sep 2016 | Details |
| NVIDIA | Quadro P6000 | Pascal | PCIe | 24 | 12.63 | 250 | Oct 2016 | Details |
| NVIDIA | Quadro RTX 6000 | Turing | PCIe | 24 | 130.5 | 295 | Aug 2018 | Details |
| NVIDIA | RTX 4500 Ada Generation | Ada Lovelace | PCIe | 24 | 97 | 210 | Mar 2023 | Details |
| NVIDIA | RTX A5000 | Ampere | PCIe | 24 | 54.2 | 230 | Apr 2021 | Details |
| Tenstorrent | Wormhole | Tensix Core | — | 24 | 364 | 160 | Oct 2023 | Details |
| Enflame Technology | CloudBlazer T20 | GCU (AI Graphics Unit) | — | 32 | — | 250 | Nov 2021 | Details |
| NVIDIA | GeForce RTX 5090 | Blackwell | PCIe | 32 | 318 | 575 | Jan 2025 | Details |
| AWS | Inferentia2 | Inferentia Gen2 | — | 32 | 190 | 150 | Nov 2022 | Details |
| Baidu | Kunlun II | Kunlun Core | — | 32 | — | 200 | Aug 2021 | Details |
| AMD | MI100 | CDNA | PCIe | 32 | 23.1 | 300 | Nov 2020 | Details |
| Moore Threads | MTT S80 | Chunxiao | — | 32 | — | 300 | Mar 2022 | Details |
| AMD | Radeon Pro V520 | RDNA 2 | PCIe | 32 | 23.04 | 225 | Jun 2021 | Details |
| NVIDIA | RTX 5000 Ada Generation | Ada Lovelace | PCIe | 32 | 130 | 250 | Mar 2023 | Details |
| NVIDIA | RTX 5000 Blackwell | Blackwell | PCIe | 32 | 242 | 250 | Jan 2025 | Details |
| TPU v4 | TPU v4 | Mezzanine | 32 | 275 | 300 | May 2021 | Details | |
| AWS | Trainium | Inferentia/Trainium | — | 32 | 190 | 200 | Nov 2021 | Details |
| NVIDIA | V100S | Volta | PCIe | 32 | 16.4 | 250 | Nov 2019 | Details |
| FuriosaAI | Warboy | FuriosaAI NPU | — | 32 | — | 80 | Oct 2021 | Details |
| Cerebras | WSE-3 | Wafer Scale Engine | — | 44 | — | 23000 | Mar 2024 | Details |
| NVIDIA | A40 | Ampere | PCIe | 48 | 150 | 300 | Oct 2020 | Details |
| Iluvatar CoreX | BI-V150 | Iluvatar GPU | — | 48 | 300 | 350 | Apr 2023 | Details |
| Intel | Data Center GPU Max 1100 | Ponte Vecchio | OAM | 48 | 177 | 300 | Jan 2023 | Details |
| NVIDIA | L40 | Ada Lovelace | PCIe | 48 | 181.05 | 300 | Oct 2022 | Details |
| NVIDIA | L40S | Ada Lovelace | PCIe | 48 | 733 | 350 | Oct 2023 | Details |
| Cambricon | MLU370 | MLU Core | — | 48 | 256 | 300 | Nov 2022 | Details |
| FuriosaAI | RNGD (Renegade) | Tensor Contraction Processor | PCIe | 48 | 256 | 180 | Jan 2024 | Details |
| NVIDIA | RTX 5880 Ada Generation | Ada Lovelace | PCIe | 48 | 167 | 250 | Mar 2024 | Details |
| NVIDIA | RTX 6000 Ada Generation | Ada Lovelace | PCIe | 48 | 182.5 | 300 | Sep 2022 | Details |
| NVIDIA | RTX A6000 | Ampere | PCIe | 48 | 77.9 | 300 | Oct 2020 | Details |
| NVIDIA | A16 | Ampere | PCIe | 64 | 71.2 | 250 | Apr 2021 | Details |
| Huawei | Ascend 910B | Da Vinci | — | 64 | — | 400 | Aug 2023 | Details |
| Rebellions | ATOM | Rebellions NPU | — | 64 | — | 200 | Jun 2023 | Details |
| Biren Technology | BR100 | Biren GPU | — | 64 | — | 550 | Aug 2022 | Details |
| Microsoft | Maia 100 | Maia | — | 64 | 700 | 500 | Nov 2023 | Details |
| AMD | MI210 | CDNA 2 | PCIe | 64 | 181 | 300 | Jan 2022 | Details |
| NVIDIA | A100 SXM | Ampere | SXM | 80 | 312 | 400 | May 2020 | Details |
| NVIDIA | A800 | Ampere | SXM | 80 | 312 | 400 | Nov 2022 | Details |
| NVIDIA | H100 SXM | Hopper | SXM | 80 | 1,979 | 700 | Sep 2022 | Details |
| TPU v5p | TPU v5 | Mezzanine | 95 | 459 | 400 | Aug 2023 | Details | |
| Intel Habana | Gaudi 2 | Gaudi Gen2 | OAM | 96 | 432 | 600 | May 2022 | Details |
| NVIDIA | GH200 | Hopper | SXM | 96 | 989 | 1000 | May 2023 | Details |
| NVIDIA | RTX 6000 Blackwell | Blackwell | PCIe | 96 | 362 | 300 | Jan 2025 | Details |
| AWS | Trainium2 | Trainium Gen2 | — | 96 | 680 | 300 | Nov 2023 | Details |
| Intel | Data Center GPU Max 1550 | Ponte Vecchio | OAM | 128 | 419 | 600 | Jan 2023 | Details |
| Intel Habana | Gaudi 3 | Gaudi Gen3 | OAM | 128 | 1,835 | 900 | Apr 2024 | Details |
| AMD | MI250X | CDNA 2 | OAM | 128 | 383 | 560 | Nov 2021 | Details |
| Meta | MTIA v1 | Meta Training & Inference Accelerator | — | 128 | — | 400 | May 2023 | Details |
| NVIDIA | H200 SXM | Hopper | SXM | 141 | 1,979 | 700 | Nov 2023 | Details |
| NVIDIA | B100 | Blackwell | SXM | 192 | 1,800 | 700 | Mar 2024 | Details |
| NVIDIA | B200 | Blackwell | SXM | 192 | 2,250 | 1000 | Mar 2024 | Details |
| NVIDIA | GB200 | Blackwell | SXM | 192 | 2,500 | 1000 | Mar 2024 | Details |
| AMD | MI300X | CDNA 3 | OAM | 192 | 1,307 | 750 | Dec 2023 | Details |
| Groq | LPU Inference Engine | TSP (Tensor Streaming Processor) | — | 230 | — | 300 | Feb 2024 | Details |
| AMD | MI325X | CDNA 3.5 | OAM | 256 | 1,400 | 800 | Oct 2024 | Details |
| AMD | MI350X | CDNA 4 | OAM | 288 | 2,100 | 850 | Jun 2025 | Details |
| AMD | MI355X | CDNA 4 | OAM | 288 | 5,300 | 1000 | Jun 2025 | Details |
| SambaNova | SN40L | Reconfigurable Dataflow Unit (RDU) | — | 640 | — | 700 | Jan 2024 | Details |
128
Total XPUs
23
Vendors
6
Benchmarks
471
Price Points