Skip to main content

AI Compute Card Encyclopedia

AI Compute Card Encyclopedia

A comprehensive catalog of 160+ AI compute cards covering training, inference, and edge computing across all scenarios. Spanning 14 vendors including NVIDIA, AMD, Intel, Google, AWS, Huawei, Cambricon, Cerebras, and more.

Quick Start

FeatureDescriptionLink
Full Comparison TableSide-by-side comparison of 160+ compute cards with filteringView Table
Future Roadmap2025–2028 product launch timelineView Roadmap
TCO CalculatorTotal cost of ownership estimation (power/PUE/NPV)Use Calculator
Industry NewsLatest releases, acquisitions, IPOs, and in-depth analysisBrowse Blog

Browse by Type

TypeDescriptionRepresentative Products
Datacenter Training GPUPrimary LLM training hardwareH100 / H200 / B200 / MI300X
Datacenter Inference GPUHigh-throughput online inferenceL4 / L40S / RTX 6000 Ada
Training ASICDedicated training acceleratorsTPU v6e / Trainium 3 / Ascend 950
Inference ASICDedicated inference acceleratorsTPU 8i / Inferentia 2 / Ascend 310
Wafer-Scale TrainingSingle-wafer solutions for massive modelsCerebras WSE-3 / WSE-4
Workstation GPUProfessional graphics + AI inferenceRTX 6000 Ada / RTX Pro 6000
Consumer GPUPersonal AI development & inferenceRTX 5090 / RTX 5080
Edge AIOn-device inference accelerationJetson Orin / Edge TPU
Innovative ArchitecturesNon-traditional architecture explorationGroq LPU / Graphcore IPU

Browse by Vendor

VendorFlagship Product LineRepresentative Models
NVIDIAGeneral-purpose GPU + CUDA ecosystemH100 / B200 / Rubin R200
AMDInstinct series + ROCmMI300X / MI400 / MI455X
Intel (Habana)Gaudi series + OpenVINOGaudi 3 / Jaguar Shores (Cancelled)
GoogleTPU tensor processorsTPU v6e / TPU 8t / TPU 8i
AWSTrainium / InferentiaTrainium 3
CerebrasWafer-Scale Engine (WSE)WSE-3 / WSE-4 🔮
HuaweiAscend series (domestic China)Ascend 910C / 920 / 950DT
CambriconSiyuan MLU seriesMLU370 / MLU690
MetaMTIA inference acceleratorMTIA v3
MicrosoftMaia inference acceleratorMaia 200
BaiduKunlun XPUKunlun R200
Moore ThreadsMTT series (domestic GPU)MTT S5000
MetaXXiyun series (domestic GPU)C600
Iluvatar CoreXTian-gai series (domestic GPU)BI-V150
EnflameT-series (domestic DTU)T21

Status: ✅ Released | 🔄 Upcoming | 🔮 Forward-looking | ❌ Cancelled

Latest Releases

Why MirrorFrog?

  1. Comprehensive — 160+ compute cards, 14 vendors, complete specifications
  2. Continuously Updated — Keeping pace with vendor release cadence, 2025–2026 launches updated promptly
  3. Neutral Comparison — No vendor bias, purely data-driven
  4. Open SourceOpen source on GitHub, PRs welcome
  5. TCO Tool — Built-in Total Cost of Ownership calculator with power/PUE/NPV estimation

Contribute

Found an error or want to add a new card? Open an Issue or submit a PR!


Last updated: 2026-07-22 | ← Home | Comparison Table → | Roadmap → | TCO Calculator → | Blog →