Cerebras

Democratizing large AI Models without HPC scaling requirements.

The AMD vs Nvidia GPU battle rages on. Here's how the two stack up.

The future for quantum computing seems to first pass through an increasing entanglement with High-Performance Computing (HPC) facilities. The future isn't just quantum: it's hybrid.

CEA, a French research institution, and C12, a quantum computing specialist, have announced they expect to tape out a million-qubit, wafer-scale chip by 2024.

A render of a Bitcoin mining rig promising 440 TH/s poses as Cerebras' CS-2 system, packing the world's largest semiconductor chip at 850,000 cores. But is this a real product or a PR stunt?

Cerebras announced that it has enabled brain-scale AI models with a new clustering technique that enables up to 163 million cores in a single cluster.

A forward-looking solution in the computing route may include wafer-sized chips packing a number of different IP blocks.

Cerebras shared more details about its Wafer Scale Engine 2, which comes with 850,000 AI cores.

Cerebras teased its new Wafer Scale Engine with 2.6 trillion 7nm transistors and 850,000 cores at Hot Chips 2020.

Cerebras' world's largest chip takes compute to a whole new level.