#cerebras — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #cerebras, aggregated by home.social.
-
От архитектуры RISC-V в ЦОД до 256-ядерных чипов Diamond Rapids: новинки серверного железа за август 2026
Август 2026 года выдался насыщенным с точки зрения новинок. Hot Chips 2026 и Flash Memory Summit принесли целую волну анонсов железа: от рекордных SSD до процессоров с гигабайтами кэша, а еще все, что раньше считалось памятью, теперь становится хранилищем, и наоборот. Собрали самое интересное из того, что появилось на рынке в августе, в традиционном дайджесте
https://habr.com/ru/companies/selectel/articles/1082542/
#selectel #процессоры #ssd #cxl #intel #amd #riscv #supermicro #cerebras #серверы
-
RT @imnotchalk: Qwen3.8 27B ist jetzt im @cerebras Shared Tier verfügbar mit einer Geschwindigkeit von ~1500 Token pro Sekunde. 🚀
mehr auf Arint.info
#AI #Cerebras #LLM #MachineLearning #Qwen3 #TechNews #arint_info
-
🚀🎉 Big news, everyone! The latest and greatest Qwen 3.8 27B is out on #Cerebras, promising to process text faster than you can say "overhyped tech jargon." Just navigate through a labyrinth of buzzwords and obscure documentation to find out how to pay for it - because who doesn't love a treasure hunt through a digital maze? 💸🔍
https://inference-docs.cerebras.ai/models/overview #Qwen3.8 #TechBuzz #DigitalMaze #TreasureHunt #HackerNews #ngated -
Qwen 3.8 27B available on Cerebras at 1500 tok/SEC
https://inference-docs.cerebras.ai/models/overview
Comments: https://news.ycombinator.com/item?id=49554520
#HackerNews #Qwen3.8 #Cerebras #AI #Model #1500tok #SEC #MachineLearning
-
At Hot Chips 2026, wafer scale engine developer Cerebras is talking about the next generation of their giant accelerators, as well as how the company is going rack-scale thanks to their Nexus platform architecture#AI #Cerebras #HotChips2026
Cerebras Talks Going Rack-Scale with Their WSEs at Hot Chips 2026 -
Cerebras is taking a rack-scale approach to AI inference with the CS-4, combining three wafer-scale engines and faster I/O instead of filling racks with separate GPUs.
-
Cerebras this week has introduced a major upgrade to its hardware ecosystem. The company is launching their first rack-scale AI inference system, the CS-4, which is powered by the upgraded WSE-3 Turbo processor#AI #Cerebras
Cerebras Intros Faster WSE-3 Turbo Processor and First Rack-Scale CS-4 System