Sophon PFG-1: a monolithic-3D AI ASIC with 330 GB of on-die DRAM and no HBM
Key takeaways
- PFG-1 "Sophon" is a unified training-and-inference die on a 750 mm², 32-tier 2D Transition-Metal Dichalcogenide (TMD) Monolithic 3D (M3D) platform.
- Compute is pure digital Compute-In-Memory (CIM): each 256×256 DRAM subarray tile pairs a binary sense amplifier with an 8-level adder tree, driven by a 500 MHz bit-serial activation broadcast.
- Sophon eliminates off-die High-Bandwidth Memory (HBM) entirely.
PFG-1 "Sophon" is a unified training-and-inference die on a 750 mm², 32-tier 2D Transition-Metal Dichalcogenide (TMD) Monolithic 3D (M3D) platform. Weights, gradients, and optimizer state reside in on-die 2T0C 2D-TMD gain-cell DRAM; because the array is fully read-write, the same silicon executes BF16 forward/backward training passes and serves low-batch decode at the compute-bound rate.
Compute is pure digital Compute-In-Memory (CIM): each 256×256 DRAM subarray tile pairs a binary sense amplifier with an 8-level adder tree, driven by a 500 MHz bit-serial activation broadcast. At 131,072 tiles/die this yields 4,200 TFLOPS FP8 and 2,100 TFLOPS BF16 in a 7.5 cm² footprint.
The die is built on a 28 nm Si Complementary Metal-Oxide-Semiconductor (CMOS) base tier, a 32-tier 2D-TMD CMOS MAC stack, and a Monolithic Inter-tier Via (MIV) fabric [5][6][7], with the 2T0C DRAM module embedded at the Back-End-Of-Line (BEOL) Metal-3 layer of each memory tier. The die stack cross-section is shown in Figure 1.