Material Insights · Understanding the Material World

AI accelerator chip: Evolution

Timeline

The milestones of the product, each with a source

YearEvent
1989Intel announces the 80170 ETANN, an analogue neural-network chip with 64 neurons and 10,240 synapses; no sale is recorded in the pages opened [25][26].
2015Google's tensor processing unit (TPU) is deployed in its own data centres [27].
2016NVIDIA's Tesla P100, with HBM2 memory stacks in the same package (CoWoS), is bought in bulk by hyperscalers, and the first DGX-1 ships [52][53].
2017NVIDIA announces the Tesla V100, with 21.1 billion transistors, 640 tensor cores, 16 GB of memory on the module and a 300 W design power; shipping was set for the third quarter [28].
2024NVIDIA announces Blackwell with 208 billion transistors on two reticle-limit dies [14].
2025Google's seventh-generation TPU, Ironwood, arrives with 192 GB of HBM3e memory and liquid cooling [39][40]; Gartner counts more than USD 200 billion of AI processor sales in the year [36].
2026NVIDIA reports USD 193.7 billion of data-centre revenue for its fiscal year [5]; India approves Semicon 2.0 and has onboarded over 38,000 GPUs under the IndiaAI Mission [24][23].

as of 2026-11-16

Lineage

The products and technologies it evolved from and into

Evolved from
  • GPU Graphics processing unit The data-centre GPU grew from the graphics chip: the Tesla V100 of 2017 added 640 tensor cores to a GPU die [28].
  • Integrated circuit (chip) An early neural-network chip, Intel's ETANN of 1989, was an integrated circuit with 64 analogue neurons [25].

as of 2026-11-16

Recipe changes

How what it is made of changed over time, and what drove each change

ChangeFromToDriverSource
Memory on the package16 GB of memory on the Tesla V100 module, with an HBM2 interface (2017)80 GB of HBM3 on the H100 SXM and 192 GB of HBM3e on Google's IronwoodLarge models need more weights held close to the arithmetic units [28][4][39][28][4][39]
Number of logic diesOne die of 815 mm2 (V100)Two reticle-limit dies joined by a 10 TB/s link (Blackwell)A single die cannot grow past the reticle limit of 858 mm2 [28][14][43][28][14][43]
Power per package300 W (V100, 2017)700 W (H100 SXM) and 750 W (AMD MI300X)More transistors and memory run faster, and cooling must remove the heat [28][4][51][28][4][51]
CoolingAir cooling in the DGX H100Liquid cooling in Google's IronwoodNot stated in the pages opened; Google says Ironwood gives 2x the performance per watt of its previous chip [40][21][40]

as of 2026-11-16

What could improve it next

Directions for improvement, today's value and the limit

DirectionMetricTodayLimitUnitSource
Less embodied carbon per unit of computeEmbodied carbon intensity of Google TPUs38– gCO2e per exaflop (TPU v6e; 114 for TPU v4i)[3]
Stack chips in 3D instead of spreading them out in 2.5DRise in embodied emissions when compute units and HBM capacity are doubled14189% increase (today 2.5D; limit shown is 3D)[11]
Bigger interposers and more HBM stacks per packageInterposer sizetwo reticle-limit dies per GPU14reticle sizes (TSMC 2029 roadmap)[43][14]
More work per wattPerformance per watt of Google's Ironwood against its predecessor2– times[40]

as of 2026-11-16

Sources

  1. Life-cycle emissions of AI hardware: a cradle-to-grave approach and generational trends Schneider et al. (Google), arXiv preprint 2502.01671, 2025
  2. NVIDIA H100 Tensor Core GPU: specifications NVIDIA, 2026
  3. NVIDIA Announces Financial Results for Fourth Quarter and Fiscal 2026 NVIDIA investor relations, 2026
  4. 3D Integration Reduces Embodied Emissions Impact Of GPU Scaling (imec and VUB) Semiconductor Engineering, 2026
  5. NVIDIA Blackwell Platform Arrives to Power a New Era of Computing NVIDIA investor relations, 2024
  6. Introduction to NVIDIA DGX H100/H200 Systems (last updated 17 October 2026) NVIDIA, 2026
  7. IndiaAI Mission expands AI ecosystem with affordable compute and startup support Digital India, Government of India, 2026
  8. Cabinet approval of Semicon 2.0 (Prime Minister's Office page) Prime Minister's Office, Government of India, 2026
  9. Neural Network Chip Joins the Collection Computer History Museum, 2024
  10. John C. Dvorak on Intel's First Neural Network Chip The Chip Letter (Substack), 2024
  11. In-Datacenter Performance Analysis of a Tensor Processing Unit (Jouppi et al.) Google Research, 2017
  12. NVIDIA Announces Tesla V100 with Volta GPU at GTC 2017 PC Perspective, 2017
  13. By decade's end, AI will drive more than half of all chip sales (reporting Gartner) The Next Platform, 2026
  14. Google's Ironwood TPUs represent a bigger threat than Nvidia would have you believe The Register, 2025
  15. Google offers 7th-gen Ironwood TPUs for AI, with AI-inspired comparisons The Register, 2025
  16. TSMC's next-gen CoWoS roadmap: over 14-reticle packages and a 48x leap in compute power expected by 2029 Tom's Hardware, 2026
  17. AMD unveils Instinct MI300X GPU and MI300A APU, claims up to 1.6X lead over Nvidia's competing GPUs Tom's Hardware, 2023
  18. NVIDIA Tesla P100: technical overview NVIDIA, 2016
  19. Deep Learning Drives Nvidia's Tesla Business To New Highs The Next Platform, 2016