Cerebras CS-4 Hits 30x Faster Inference Without Building a New Chip
Cerebras unveils CS-4 with three overclocked WSE-3 Turbo wafers, delivering 750 PFLOPS and up to 30x faster inference than GPU systems.
- Cerebras launched CS-4, a rack-scale AI system with three WSE-3 Turbo wafers delivering 750 PFLOPS
- Claims up to 30x faster inference than GPUs and 10x more throughput per watt than CS-3
- WSE-3 Turbo is not new silicon, just the WSE-3 overclocked with twice the power delivered
- New Nexus architecture uses modular wafer backpacks with 50% fewer components and 2-microsecond wafer-to-wafer latency
- Targets frontier inference: 1,000+ tokens per second on 10T-parameter models for agentic workloads
- First shipments begin this quarter; see the official blog for details
Cerebras just pulled back the curtain on CS-4, its next rack-scale AI system, and the headline numbers are aggressive: up to 30x faster inference than GPU systems and up to 10x more throughput per watt than the CS-3 it replaces. What is interesting is how they got there. Instead of shipping a brand new silicon generation, Cerebras built a new rack architecture around an overclocked version of its existing wafer, and squeezed twice the work out of the same chip.
What is actually inside the box
The Cerebras CS-4 is a rack-scale AI accelerator built from three WSE-3 Turbo wafers, delivering 750 PFLOPS of compute. It is the first system on the Nexus platform and targets ultrafast inference for very large AI models. The full spec sheet reads like a bandwidth flex:
- 750 PFLOPS of AI compute, 129.6 petabytes per second of memory bandwidth, and 7.2 terabits per second of system I/O bandwidth
- Wafer-to-wafer latency as low as two microseconds to support very large AI models
- Up to 10x more throughput per watt than CS-3
- Up to 2x faster than CS-3 and up to 30x more tokens-per-second-per-user than leading GPU solutions
Each of the three wafers is a WSE-3 Turbo. Like the WSE-3, the WSE-3T is the largest AI processor ever built, containing four trillion transistors and 900,000 AI-optimized cores across 46,225 square millimeters of silicon, with 44GB of SRAM integrated directly on the wafer. The WSE-3T doubles AI compute to 250 PFLOPS per wafer and doubles memory bandwidth to 43.2 petabytes per second.
The trick: same silicon, twice the juice
Here is the part worth understanding. The WSE-3T is not a new chip. The CS-4 machines are getting an overclocked version of the current WSE-3 waferscale compute engine, with the exact same 900,000 cores and the exact same 44 GB of on-wafer SRAM, and made using the same 5 nanometer processes. If you look at the chart, you'll notice it accomplishes this using the same process tech, wafer area size, transistor count, core count, and SRAM capacity. That's because the WSE-3T isn't new silicon. Instead, Cerebras tells us it's just pushing its existing wafer scale engine harder.