Evidence96%Authoritative
FactConfirmedProduct·November 6, 2025
CS-3 vs. NVIDIA DGX B200 Blackwell Benchmarks Published
Cerebras published benchmarks showing the CS-3 is 21x faster at inference than NVIDIA's DGX B200 Blackwell on Llama 3 70B, with 1/3 the cost and 1/3 the power, and maintained a ~5x speed advantage on GPT-OSS-120B against Blackwell's best GPU results.
Evidence Strength
Evidence96%Authoritative
Backed by official company doc
Single publisher source
Includes official or primary source
Key Development
High-significance development (rated 7/10)
Covered by 3 sources
Confirmed — verified event
Insights
First tracked
September 19, 2025
Last updated
November 6, 2025
Sources
3 sources
Related Developments
Oklahoma City AI Datacenter Ribbon-Cutting with 44+ ExaflopsCerebras Delivers 3,000 Tokens/Second Inference for OpenAI's gpt-oss-120B Open-Weight ModelJais 2 Arabic-Centric LLMs Trained and Deployed on Cerebras Wafer-Scale ClustersGLM-4.7 Available on Cerebras Inference Cloud at 1,000-1,700 Tokens/SecondOpenAI Signs $10B+ Multiyear Compute Deal with Cerebras
Sources (3)
Source Timeline
OpenAI GPT-OSS 120B Benchmarked – NVIDIA Blackwell vs. CerebrasCerebras·Nov 6, 2025- Cerebras CS-3 vs. Groq LPUCerebras·Sep 19, 2025
- Cerebras CS-3 vs. Nvidia DGX B200 BlackwellCerebras·Sep 19, 2025
Evidence Strength
Evidence96%Authoritative
Backed by official company doc
Single publisher source
Includes official or primary source
Key Development
High-significance development (rated 7/10)
Covered by 3 sources
Confirmed — verified event
Insights
First tracked
September 19, 2025
Last updated
November 6, 2025
Sources
3 sources
Related Developments
Oklahoma City AI Datacenter Ribbon-Cutting with 44+ ExaflopsCerebras Delivers 3,000 Tokens/Second Inference for OpenAI's gpt-oss-120B Open-Weight ModelJais 2 Arabic-Centric LLMs Trained and Deployed on Cerebras Wafer-Scale ClustersGLM-4.7 Available on Cerebras Inference Cloud at 1,000-1,700 Tokens/SecondOpenAI Signs $10B+ Multiyear Compute Deal with Cerebras