aiqbench takeChips & Infra

Semifive has successfully transitioned its AI business model from one-off projects to sustained revenue streams by mass-producing the 'Bertha' inference accelerator on Samsung's 4nm process. This chip specifically targets data center workloads for large language model inference, marking a significant step in the industry's shift away from training-focused hardware toward scalable inference solutions.

Source coveringQuantum Zeitgeist

What this isA model's reading of the article and related coverage, written for aiqbench. Not the publisher's words.

Semifive starts shipping AI chips built on Samsung’s 4nm process

Quantum ZeitgeistSeptember 8, 20262 min read

SEMIFIVE announced today the mass production of a data center AI inference accelerator for HyperAccel, marking its first large-scale project utilizing Samsung Foundry’s 4nm process node. Following an initial production contract, mass production volumes are projected to expand with follow-on purchase orders aligned with HyperAccel’s rollout.

SEMIFIVE has begun mass production of the ‘Bertha’ AI inference accelerator for HyperAccel, a data center chip exceeding 500 mm² in size and built on Samsung’s 4nm process. This achievement represents SEMIFIVE’s third mass-production…

Excerpt shown under the publisher's RSS terms · 34 words · full text and figures are on the publisher's site

Read at Quantum Zeitgeist ↗Ask about this article