by SemiAnalysis
SemiAnalysis's rating system that ranks GPU cloud (neocloud) providers, with CoreWeave and Nebius highlighted.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
ClusterMAX 3.0 is the third edition of SemiAnalysis's rating system for GPU cloud providers, often called neoclouds. It ranks providers on how they are tested and evaluated, and publishes tiered ratings. According to a YouTube discussion of the report, CoreWeave and Nebius were highlighted in the rankings, and the report drew some backlash from providers.
It is published by SemiAnalysis, an independent research firm covering AI and semiconductor supply chains. The provided pages describe SemiAnalysis's broader products (accelerator and HBM model, datacenter model, tokenomics model, consulting, and a GPU Pricing API) but give no detail on ClusterMAX methodology, criteria or full ratings, so those are not described here.
Access is through SemiAnalysis's website and publications. Some SemiAnalysis content is paywalled for subscribers, but the provided material does not confirm which parts of ClusterMAX are free or paid.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
SemiAnalysis's rating system for GPU cloud (neocloud) providers. Chapters cover how clouds are ranked, how providers are tested, the ratings, and the backlash, with CoreWeave and Nebius highlighted in the rankings.
Closest neighbours by meaning, not just by category.
Model · NVIDIA
NVIDIA's open-weight 30B mixture-of-experts model with 1M-token context, aimed at long-running agent workloads.
Model · NVIDIA
NVIDIA family of multilingual text embedding models (8B and 1B) for retrieval, RAG and coding agents.
Research · MiniMax
MiniMax's open-source sparse top-k and dense attention GPU kernels for NVIDIA SM100 (Blackwell), with a technical report.
Library
Windows-wheel fork of SageAttention, a quantized attention kernel library that speeds up diffusion and video model inference.
Research · Artificial Analysis
Artificial Analysis's aggregate benchmark index that scores language models on intelligence across multiple evaluations.
Model · NVIDIA
NVIDIA research system where a code game engine drives a two-step streaming MiniMax-H3 video model for real-time rendering.