BTC $64,654.85 -0.96%
ETH $1,812.37 -0.79%
BNB $593.02 -0.13%
XRP $0.9734 +1.83%
SOL $74.01 +3.50%
TRX $0.3075 +0.81%
DOGE $0.0596 +1.02%
ADA $0.1544 +0.78%
BCH $210.85 +0.72%
LINK $9.25 +1.30%
HYPE $69.20 -1.12%
AAVE $111.75 +3.63%
SUI $0.7328 +1.68%
XLM $0.1399 +1.67%
ZEC $947.00 +3.87%
AAPL $351.47 -0.71%
AMZN $261.22 +1.02%
GOOGL $283.52 -0.32%
MSFT $511.81 -1.66%
META $624.95 -0.11%
NVDA $162.34 +3.96%
TSLA $384.83 +1.33%
SNDK $841.96 +4.01%
INTC $56.39 +5.36%
SPCX $142.78 +3.55%
MU $746.33 +3.81%
AMD $522.77 +3.59%
BTC $64,654.85 -0.96%
ETH $1,812.37 -0.79%
BNB $593.02 -0.13%
XRP $0.9734 +1.83%
SOL $74.01 +3.50%
TRX $0.3075 +0.81%
DOGE $0.0596 +1.02%
ADA $0.1544 +0.78%
BCH $210.85 +0.72%
LINK $9.25 +1.30%
HYPE $69.20 -1.12%
AAVE $111.75 +3.63%
SUI $0.7328 +1.68%
XLM $0.1399 +1.67%
ZEC $947.00 +3.87%
AAPL $351.47 -0.71%
AMZN $261.22 +1.02%
GOOGL $283.52 -0.32%
MSFT $511.81 -1.66%
META $624.95 -0.11%
NVDA $162.34 +3.96%
TSLA $384.83 +1.33%
SNDK $841.96 +4.01%
INTC $56.39 +5.36%
SPCX $142.78 +3.55%
MU $746.33 +3.81%
AMD $522.77 +3.59%

cs-4

All
Article
Flash

hot_img Cerebras releases the fourth generation AI inference system CS-4: performance doubled, power consumption doubled, more flexible deployment

Cerebras released its fourth-generation AI inference system CS-4 this week, based on the same 5nm WSE-3 wafer, achieving double the performance by doubling the clock frequency and power consumption. A single CS-4 cabinet accommodates 3 wafers (CS-3 has 2), featuring a modular "backpack" design that simplifies manufacturing and deployment, with a TDP of approximately 125 to 135kW. The CS-4 can provide an inference speed of nearly 4000 tokens/second/user, about twice that of the CS-3, and supports decomposed inference with heterogeneous systems such as AMD and AWS Trainium.Cerebras claims that the CS-4 offers about 2000 times the on-chip memory bandwidth of NVIDIA's Rubin (43PB/s), but the 44GB SRAM capacity remains unchanged, and long-context inference still requires multi-wafer stacking. For example, with the DeepSeek V4 Pro (1.6T parameters), approximately 20 systems are needed for a 1M context window, and about 40 systems are required for 256 concurrent users, corresponding to a CAPEX exceeding 20 million USD. Cerebras is collaborating with clients such as OpenAI and plans to achieve approximately double performance improvements each year, aiming for a 20-fold throughput increase by 2027. The "backpack" cabinet design of the CS-4 will continue into the next-generation "Nexus" platform.
app_icon
ChainCatcher Building the Web3 world with innovations.