InferenceMAX™: Open Source Inference Benchmarking

NVIDIA GB200 NVL72, AMD MI355X, Throughput Token per GPU, Latency Tok/s/user, Perf per Dollar, Tokens per Provisioned Megawatt, DeepSeek R1 670B, GPTOSS…

Oct 9 • 

Kimbo Chen

, 

Dylan Patel

, 

Daniel Nishball

, 

Cam Quilici

, and 

Cheang Kang Wen

106

4

September 2025

xAI's Colossus 2 - First Gigawatt Datacenter In The World, Unique RL Methodology, Capital Raise

On Site Turbines, Mississippi Expansion, Solaris Energy, Can xAI afford it?, Middle East Funding, Tesla, Talent Exodus, API revenue, Consumer Growth, RL…

Sep 16 • 

Jeremie Eliahou Ontiveros

, 

Dylan Patel

, 

Wei Zhou

, 

AJ Kourabi

, and 

Maya Barkin

14

Another Giant Leap: The Rubin CPX Specialized Accelerator & Rack

New Prefill Specialized GPU, Rack Architecture, BOM, Disaggregated PD, Higher Perf per TCO, Lower TCO, GDDR7 & HBM Market Trends

Sep 10 • 

Dylan Patel

, 

Daniel Nishball

, 

Kimbo Chen

, 

Myron Xie

, 

Wega Chu

, 

Gerald Wong

, 

Cheang Kang Wen

, and 

Ivan Chiam

4

Huawei Ascend Production Ramp: Die Banks, TSMC Continued Production, HBM is The Bottleneck

H20 Shipments, Blackwell B30A, Bottlenecks to Chinese Chip Production, Export Controls, CXMT, SMIC, Cambricon

Sep 8 • 

Dylan Patel

, 

AJ Kourabi

, 

Myron Xie

, and 

Jeff Koch

1

Amazon’s AI Resurgence: AWS & Anthropic's Multi-Gigawatt Trainium Expansion

Anthropic multi-gigawatt clusters, Trainium ramp, best TCO per memory bandwidth, system-level roadmap, Bedrock and internal models

Sep 3 • 

Jeremie Eliahou Ontiveros

, 

Dylan Patel

, 

AJ Kourabi

, and 

Myron Xie

August 2025

H100 vs GB200 NVL72 Training Benchmarks - Power, TCO, and Reliability Analysis, Software Improvement Over Time

Joules per Token, TCO Per Million Tokens, MFU, Tokens Per US Annual Household Energy Usage, DeepSeek 670B, GB200 Unreliability, Backplane Downtime

Aug 20 • 

Dylan Patel

 and 

Daniel Nishball

GPT-5 Set the Stage for Ad Monetization and the SuperApp

How ChatGPT will monetize free users, Router is the Release, AIs will serve Ads, Google's moat eroded?, The shift of purchasing intent queries

Aug 13 • 

Doug

, 

Dylan Patel

, 

Wei Zhou

, and 

AJ Kourabi

1

Scaling the Memory Wall: The Rise and Roadmap of HBM

HBM4, Custom Base Die, Shoreline Expansion, Process Flow, China Domestic Production, Samsung Qualification

Aug 12 • 

Dylan Patel

, 

Myron Xie

, 

Tanj Bennett

, 

Ivan Chiam

, and 

Jeff Koch

July 2025

Robotics Levels of Autonomy

Single-Purpose Robots Automating Hundreds Of Jobs, Pick And Place With Low Autonomy Is Expensive, General-Purpose Autonomy Navigating And Inspecting…

Jul 30 • 

Reyk Knuhtsen

, 

Dylan Patel

, 

Jeremie Eliahou Ontiveros

, 

Joe Ryu

, 

Robert Ghilduta

, and 

Niko

1

Intel 18A Details & Cost, Future of DRAM 4F2 vs 3D, Backside Power Adoption (or Not), China's FlipFET, Digital Twins from Atoms to Fabs, and…

VLSI 2025 Roundup

Jul 21 • 

Dylan Patel

, 

Jeff Koch

, and 

Gerald Wong

Meta Superintelligence - Leadership Compute, Talent, and Data

AI Datacenter Titanomachy, "The Tent", AI Data and Talent Wars, Zuck Founder Mode, Behemoth 4 Post-Mortem, OBBB Tax Windfall, AI and Reality Labs

Jul 11 • 

Dylan Patel

, 

Jeremie Eliahou Ontiveros

, 

Wei Zhou

, 

AJ Kourabi

, and 

Maya Barkin

2

DeepSeek Debrief: >128 Days Later

Traffic and User Zombification, GPU Rich Western Neoclouds, Token Economics (Tokenomics) Sets the Competitive Landscape

Jul 3 • 

Wei Zhou

, 

AJ Kourabi

, and 

Dylan Patel

2