During the AI Infra Summit, NVIDIA announced the "Rubin CPX" GPU, a specialized accelerator derived from the upcoming "Rubin" family, specifically made for massive-context AI models. The chip delivers 30 PetaFLOPS of NVFP4 compute performance on a monolithic die, accompanied by 128 GB of GDDR7 memory. The monolithic die configuration represents a departure from the dual-GPU packages characteristic of NVIDIA's current Blackwell and Blackwell Ultra architectures, as well as the design path that the rest of the Rubin family will follow. The Rubin CPX addresses computational bottlenecks in extended-context scenarios where AI models process millions of tokens simultaneously. This capability proves critical for applications including comprehensive software codebase analysis and hour-long video content processing, which can require up to one million tokens.

The processor integrates four NVENC and four NVDEC video encoders directly on-chip, enabling streamlined multimedia workflows without external processing dependencies. Performance metrics suggest that the Rubin CPX is delivering three times the attention processing speed of NVIDIA's current best GB300 Blackwell Ultra accelerator systems. The architecture employs a cost-optimized single-die approach, rather than multi-chip modules, which potentially reduces manufacturing complexity while maintaining computational density. Memory bandwidth specifications remain undisclosed, though a 512-bit interface could yield approximately 1.8 TB/s throughput when using 30 Gbps GDDR7 memory chips. NVIDIA plans integration of Rubin CPX processors within the Vera Rubin NVL144 CPX platform, which combines traditional Rubin GPUs with the specialized CPX variants. This hybrid configuration targets 8 ExaFLOPS of aggregate compute performance, with 1.7 PB/s of memory bandwidth, across a complete rack deployment. The "Kyber" rack will include ConnectX-9 network adapters capable of 1600G networking, Spectrum6 doing 102.4T switching, as well as co-packaged optics. NVIDIA plans this to arrive in late 2026, after the regular Rubin GPU launch in early 2026.NVIDIA pushes Rubin CPX as a one-off in the Rubin family to address the complexity of inferencing a test-time scaling AI system. As models evolve beyond simple text generation toward sophisticated reasoning systems, inference operations increasingly split between computationally intensive context processing and memory-bandwidth-dependent token generation phases. The CPX design optimizes for these dual requirements through its dedicated architecture, handling context prefill operations that can span enterprise chatbot sessions with 256,000 tokens or comprehensive code analysis exceeding 100,000 lines. This specialization becomes critical as AI systems transition from basic language models to multistep reasoning agents that maintain persistent memory across extended interactions. Hence, all of that needs to be enabled by hardware, and NVIDIA wants to make it a seamless experience for developers.