Cerebras & Core42 Announce Record-Breaking Global Launch With OpenAI’s GPT-OSS-120B

Cerebras and Core42 announced the global launch of OpenAI’s GPT-OSS-120B on the Core42 AI Cloud, delivering record-breaking inference performance of 3,000 tokens per second. Powered by Cerebras’ CS-3 system and wafer-scale engine, the collaboration enables real-time reasoning, long-context understanding (128K tokens), and ultra-low-latency processing at a significantly lower cost per token than GPUs. Enterprises, researchers, and governments can now build scalable, agentic AI applications through the Core42 Compass API with industry-leading efficiency. Pricing is set at $0.25 per million input tokens and $0.69 per million output tokens. The system supports semantic search, automation, code execution, and decision intelligence, marking a new benchmark in open-weight AI infrastructure. Cerebras and Core42 emphasize unmatched speed, scalability, and compliance for enterprise-scale AI innovation.

Business Wire, Cerebras Systems Press Release – Cerebras and Core42 Deliver Record-Breaking Performance for OpenAI’s gpt-oss-120B, Powering AI Innovation for Enterprises Worldwide – 28th August 2025