Cerebras Systems
Wafer-scale engine (WSE) startup pivoting to ultra-fast inference
Cerebras builds the Wafer-Scale Engine (WSE-3), a single dinner-plate-sized chip with hundreds of thousands of cores and huge on-chip SRAM, fabbed at TSMC. In 2025 it leaned hard into a fast-inference cloud service, delivering industry-leading tokens/sec on open models, and pursued a delayed IPO amid a large capacity deal with G42 in the UAE. Its differentiation is avoiding multi-chip communication bottlenecks by keeping an entire model's compute on one wafer.
Product
WSE-3 wafer-scale engine + CS-3 systems
2025 focus
High-speed inference cloud
Key customer
G42 (UAE)
Fab
TSMC (wafer-scale)
How it fits the stack
Cerebras Systems with what it depends on (above) and what it feeds (below). The figure renders as a crawlable diagram and upgrades to an interactive 3D graph as it scrolls into view.
Cerebras Systems in the AI stack. Cerebras Systems with its immediate upstream dependencies (top) and downstream dependents (bottom) in the AI value chain. Hover a node in 3D, or read the full relationships below.
Graph data (text) — 3 entities, 2 relationships
- Cerebras Systems —depends on→ TSMC (Taiwan Semiconductor Manufacturing Company)
- Cerebras Systems —competes with→ Nvidia
Context — capital, rivals, policy · · 1