Chip leaderboard
Most transistors per die
Cerebras WSE-3 (Wafer-Scale Engine 3) leads at 4,000 B. 14 chips ranked.
How many transistors are on each chip's die. Wafer-scale designs like Cerebras WSE-3 dominate by definition; chiplet packages like GB200 and MI300X come next.
| # | Chip | Value | Inputs |
|---|---|---|---|
| 1 | Cerebras WSE-3 (Wafer-Scale Engine 3) Cerebras Systems | 4,000 B | Die size: 46225 mm² |
| 2 | NVIDIA Rubin (R100) NVIDIA | 336 B | |
| 3 | NVIDIA B200 (Blackwell) NVIDIA | 208 B | |
| 4 | NVIDIA B300 (Blackwell Ultra) NVIDIA | 208 B | |
| 5 | AMD Instinct MI300X AMD | 153 B | |
| 6 | NVIDIA H100 Tensor Core GPU NVIDIA | 80 B | Die size: 814 mm² |
| 7 | NVIDIA H200 Tensor Core GPU NVIDIA | 80 B | Die size: 814 mm² |
| 8 | Biren BR100 Biren Technology | 77 B | |
| 9 | NVIDIA L40S NVIDIA | 76.3 B | |
| 10 | NVIDIA A100 Tensor Core GPU NVIDIA | 54 B | Die size: 826 mm² |
| 11 | Tesla Dojo D1 Tesla | 50 B | Die size: 645 mm² |
| 12 | NVIDIA A10 Tensor Core GPU NVIDIA | 28.3 B | Die size: 628 mm² |
| 13 | NVIDIA V100 Tensor Core GPU NVIDIA | 21.1 B | Die size: 815 mm² |
| 14 | NVIDIA T4 Tensor Core GPU NVIDIA | 13.6 B | Die size: 545 mm² |
14 chips ranked. 46 more are in our catalog but don't publish the numbers needed for this ranking.
Other rankings: Best perf-per-watt (TFLOPS/W) · Most HBM memory per chip · Highest memory bandwidth · Highest FP8 dense throughput · Highest FP16 dense throughput · Highest INT8 dense throughput · Cheapest per TFLOP at launch (list price)
See also: every chip on record · head-to-head comparisons.