Cerebras says mixed-chip inference lifts throughput 5x in early results

Cerebras says disaggregation raised throughput using the same number of its systems without reducing token-generation speed. A separate General Compute agreement is scheduled to bring Cerebras inference capacity to customers in Q1 2027.

By · Published

Cerebras says splitting inference work lifts throughput 5x on the same systems — The early result supports a mixed-hardware strategy; a separate Cerebras deal with General Compute is set to bring its inference capacity to customers in Q1…

Cerebras says disaggregation raised throughput using the same number of its systems without reducing token-generation speed. A separate General Compute agreement is scheduled to bring Cerebras inference capacity to customers in Q1 2027.