Semiconductor device fabrication
MOSFET scaling (process nodes)
020 μm – 1968 010 μm – 1971 006 μm – 1974 003 μm – 1977  1.5 μm – 1981 001 μm – 1984 800 nm – 1987 600 nm – 1990 350 nm – 1993 250 nm – 1996 180 nm – 1999 130 nm – 2001 090 nm – 2003 065 nm – 2005 045 nm – 2007 032 nm – 2009 028 nm – 2010 022 nm – 2012 014 nm – 2014 010 nm – 2016 007 nm – 2018 005 nm – 2020 003 nm – 2022
Future 002 nm ~ 2025 001 nm ~ 2027
Half-nodes Density CMOS Device (multi-gate) Moore's law Transistor count Semiconductor Industry Nanoelectronics
v t e

Thetransistor count is the number oftransistors in an electronic device (typically on a single substrate or silicondie). It is the most common measure ofintegrated circuit complexity (although the majority of transistors in modernmicroprocessors are contained incache memories, which consist mostly of the samememory cell circuits replicated many times). The rate at whichMOS transistor counts have increased generally followsMoore's law, which observes that transistor count doubles approximately every two years. However, being directly proportional to the area of a die, transistor count does not represent how advanced the corresponding manufacturing technology is. A better indication of this is transistor density which is the ratio of a semiconductor's transistor count to its die area.

As of 2023^[update], the highest transistor count in flash memory isMicron's 2 terabyte (3D-stacked) 16-die, 232-layerV-NAND flashmemory chip, with 5.3 trillionfloating-gate MOSFETs (3 bits per transistor).

The highest transistor count in a single chip processor as of 2020^[update] is that of thedeep learning processorWafer Scale Engine 2 byCerebras. It has 2.6 trillion MOSFETs in 84 exposed fields (dies) on a wafer, manufactured using TSMC's7 nm FinFET process.^{[1][2][3][4][5]}

As of 2024^[update], theGPU with the highest transistor count isNvidia'sBlackwell-based B100 accelerator, built onTSMC's custom 4NP process node and totaling 208 billion MOSFETs.

The highest transistor count in a consumer microprocessor as of March 2025^[update] is 184 billion transistors, inApple'sARM-based dual-dieM3 Ultra SoC, which is fabricated usingTSMC's3 nmsemiconductor manufacturing process.^[6]

In terms of computer systems that consist of numerous integrated circuits, thesupercomputer with the highest transistor count as of 2016^[update] was the Chinese-designedSunway TaihuLight, which has for all CPUs/nodes combined "about 400 trillion transistors in the processing part of the hardware" and "theDRAM includes about 12quadrillion transistors, and that's about 97 percent of all the transistors."^[7] To compare, thesmallest computer, as of 2018^[update] dwarfed by a grain of rice, had on the order of 100,000 transistors. Early experimental solid-state computers had as few as 130 transistors but used large amounts ofdiode logic. The firstcarbon nanotube computer had 178 transistors and was a1-bitone-instruction set computer, while a later one is16-bit (itsinstruction set is 32-bitRISC-V though).

Ionic transistor chips ("water-based"analog limited processor), have up to hundreds of such transistors.^[8]

Estimates of the total numbers of transistors manufactured:

• Up to 2014:2.9×10²¹
• Up to 2018:1.3×10^22[9][10]

This subsectionneeds additional citations forverification. Relevant discussion may be found on thetalk page. Please helpimprove this article byadding citations to reliable sources in this subsection. Unsourced material may be challenged and removed. Find sources: "Transistor count" – news ·newspapers ·books ·scholar ·JSTOR(December 2019) (Learn how and when to remove this message)

Amicroprocessor incorporates the functions of a computer'scentral processing unit on a singleintegrated circuit. It is a multi-purpose, programmable device that accepts digital data as input, processes it according to instructions stored in its memory, and provides results as output.

The development ofMOS integrated circuit technology in the 1960s led to the development of the first microprocessors.^[11] The 20-bitMP944, developed byGarrett AiResearch for theU.S. Navy'sF-14 Tomcat fighter in 1970, is considered by its designerRay Holt to be the first microprocessor.^[12] It was a multi-chip microprocessor, fabricated on six MOS chips. However, it was classified by the Navy until 1998. The4-bitIntel 4004, released in 1971, was the first single-chip microprocessor.

Modern microprocessors typically include on-chipcache memories. The number of transistors used for these cache memories typically far exceeds the number of transistors used to implement the logic of the microprocessor (that is, excluding the cache). For example, the lastDEC Alpha chip uses 90% of its transistors for cache.^[13]

Agraphics processing unit (GPU) is a specialized electronic circuit designed to rapidly manipulate and alter memory to accelerate the building of images in a frame buffer intended for output to a display.

The designer refers to thetechnology company that designs the logic of theintegrated circuit chip (such asNvidia andAMD). The manufacturer ("Fab.") refers to thesemiconductor company that fabricates the chip using itssemiconductor manufacturing process at afoundry (such asTSMC andSamsung Semiconductor). The transistor count in a chip is dependent on a manufacturer's fabrication process, with smallersemiconductor nodes typically enabling higher transistor density and thus higher transistor counts.

Therandom-access memory (RAM) that comes with GPUs (such asVRAM,SGRAM orHBM) greatly increases the total transistor count, with thememory typically accounting for the majority of transistors in agraphics card. For example,Nvidia'sTesla P100 has 15 billionFinFETs (16 nm) in the GPU in addition to 16 GB ofHBM2 memory, totaling about 150 billionMOSFETs on the graphics card.^[195] The following table does not include the memory. For memory transistor counts, see theMemory section below.

Afield-programmable gate array (FPGA) is an integrated circuit designed to be configured by a customer or a designer after manufacturing.

Semiconductor memory is an electronicdata storage device, often used ascomputer memory, implemented onintegrated circuits. Nearly all semiconductor memories since the 1970s have usedMOSFETs (MOS transistors), replacing earlierbipolar junction transistors. There are two major types of semiconductor memory:random-access memory (RAM) andnon-volatile memory (NVM). In turn, there are two major RAM types:dynamic random-access memory (DRAM) andstatic random-access memory (SRAM), as well as two major NVM types:flash memory andread-only memory (ROM).

TypicalCMOS SRAM consists of six transistors per cell. For DRAM, 1T1C, which means one transistor and one capacitor structure, is common. Capacitor charged or not^{[clarification needed]} is used to store 1 or 0. In flash memory, the data is stored in floating gates, and the resistance of the transistor is sensed^{[clarification needed]} to interpret the data stored. Depending on how fine scale the resistance could be separated^{[clarification needed]}, one transistor could store up to threebits, meaning eight distinctive levels of resistance possible per transistor. However, a finer scale comes with the cost of repeatability issues, and hence reliability. Typically, low grade 2-bitsMLC flash is used forflash drives, so a 16 GB flash drive contains roughly 64 billion transistors.

For SRAM chips, six-transistor cells (six transistors per bit) was the standard.^[295] DRAM chips during the early 1970s had three-transistor cells (three transistors per bit), before single-transistor cells (one transistor per bit) became standard since the era of 4 Kb DRAM in the mid-1970s.^[296][297] Insingle-level flash memory, each cell contains onefloating-gate MOSFET (one transistor per bit),^[298] whereasmulti-level flash contains 2, 3 or 4 bits per transistor.

Flash memory chips are commonly stacked up in layers, up to 128-layer in production,^[299] and 136-layer managed,^[300] and available in end-user devices up to 69-layer from manufacturers.