SemiAnalysis · Bryan Shan ·

Vera Rubin NVL72 inference tests show up to 7x better token throughput per MW vs. Blackwell on a 1.6T DeepSeek model, above Huang's 3x claim for 1T-3T LLMs

Jensen Sandbagging Performance Again, 2x more Annual Profit Per GigaWatt, The More you Buy, The More you Earn, AgentX, InferenceX, Extreme Co-Design

In this story Blackwell DeepSeek Jensen Huang Vera Rubin NVL72

Vera Rubin NVL72 inference tests show up to 7x better token throughput per MW vs. Blackwell on a 1.6T DeepSeek model, above Huang's 3x claim for 1T-3T LLMs

Lead Source

SemiAnalysis — Vera Rubin NVL72 inference tests show up to 7x better token throughput per MW vs. Blackwell on a 1.6T DeepSeek model, above Huang's 3x claim for 1T-3T LLMs ↗

How this story grew

Coverage · 0 Discussion · 0

Sep 15Sep 16

More

The Neuron: The Neuron

Discussion

Related stories

The Register: Nvidia says its Groq 3 LPX racks delivered 3,400 tokens per second in an Artificial Analysis benchmark running Gemma 4 31B with a 100,000-token input sequence

The Information: Nvidia is developing a Nemotron 4 model with 1T+ parameters, up from Nemotron 3 Ultra's 550B parameters but smaller than leading Chinese open models

9to5Mac: OpenAI previews Ultrafast, an API tier powered by Cerebras that runs GPT-5.6 Sol up to 14× faster and generates up to 750 output tokens per second

Reuters: DeepSeek debuts DeepSeek-V4.1-Flash, its smallest model built on its new Causal Encoder-Decoder architecture, with 552B backbone parameters and 1M-token context

Bloomberg: DeepSeek unveils an experimental multimodal version of its V4 Flash model, saying it nears the performance of Anthropic's Opus 4.8 on multimodal agentic tests

Discuss with AI

ChatGPT ↗ Claude ↗

← Back to TechSnif