Ep. 002 - InferenceX 2.0 Release (Technical Staff) | Cam Quilici, Bryan Shan, Doug O'Laughlin, Jordan Nanos
SemiAnalysis WeeklyShow Notes
InferenceX, formerly InferenceMAX: https://inferencex.com/ Article (latest): https://newsletter.semianalysis.com/p/inferencex-v2-nvidia-blackwell-vs GitHub https://github.com/SemiAnalysisAI/InferenceX Article (original) https://newsletter.semianalysis.com/p/inferencemax-open-source-inference 00:00 Introduction to InferenceX 02:52 Evolution from InferenceMAX to InferenceX 06:06 Benchmarking and Performance Insights 08:43 The Scale of Benchmarking Work 11:39 Collaboration with AMD and Nvidia 14:52 The Evolution of Inference Benchmarking 17:34 Optimizations and Their Impact 20:47 Challenges in Composability 23:51 Multi-Token Prediction Explained 26:52 Cost Implications of Optimizations 31:06 Understanding Inference Workloads and Benchmarks 33:44 Future Plans for Inference Optimization 37:16 Roadmap for New Models and Data Sets 39:03 Challenges in Benchmarking Multi-Turn and Multi-Modal Data 42:44 Experiences with AI Models and Their Limitations 48:43 Skepticism About Future AI Improvements
More from SemiAnalysis Weekly
Brad Gerstner put our 43 gigawatt forecast for 2027 on All-In, then again in his BG2 sit-down with Jensen, and called it too aggressive. The consensus counter sits near 25 GW, built on moratorium headlines and grid connection fears. Jordan Nanos (@JordanNanos) sits down with Jeremie Ontiveros (@JeremieEO) and Reyk Knuhtsen (@robotknower), who have been tracking the same capacity curve since March 2024, and it has barely moved across every model revision. So they aren't cutting it. They're raising it.
GLM 5.3 only attends to the tokens that matter. It still has to keep every one of them.
H100 rental prices went up, on a GPU that should be sliding down its depreciation curve. That curve is what lenders underwrite. Daniel Nishball (@dnishball) kicks off our new Compute Capital Markets group by following the money through Nvidia's $497B backstops, ClusterMAX as a credit signal, and whether each token is profitable. Jordan Nanos (@JordanNanos) brings the rental data. AI debt is about to pass auto loans.
Pay Anthropic $200 a month and you can pull $12,000 of Opus 5.5 tokens at API pricing. Max Kan (@maxkan) and Andrew Megalaa bought every plan, OpenAI, Meta, xAI, MiniMax, Moonshot, Z.ai, Cursor, and measured every token type to find the ceiling. The catch isn't subsidy. Anthropic still books 50 to 60% gross margins on these plans, which means the API itself is the price gouge.
Jordan (@JordanNanos) and Abhilash (@AbhilashJain17) discuss the clients SemiAnalysis consulting serves, custom projects like the inference simulator, how client work turns into research products, neocloud due diligence, and why SLAs decide who gets funded. They also discuss Feynman specs, de-specced memory, and a 20-trillion-parameter OpenAI model.