AI News Feed
Market watch
LG UltraFine Evo 32U990A-S Hits Record-Low $1,099 on Amazon, TechRadar ReportsNewegg Labor Day Deal: 32GB Kingston Fury Beast DDR4-3200 Kit Nearly Half Price at $179.99 with CodeGalaxy Z Fold 8 inner screen corners reported with excessive giveHuawei unveils Mate XT 2 tri-fold with privacy-screen edition as Apple prepares folding iPhoneWall Street Analysts Recommend Nvidia, Uber and Marvell for Long-Term GrowthLenovo, AMD and Huawei Put Local Large-Model AI at Center of New Devices
LG UltraFine Evo 32U990A-S Hits Record-Low $1,099 on Amazon, TechRadar ReportsNewegg Labor Day Deal: 32GB Kingston Fury Beast DDR4-3200 Kit Nearly Half Price at $179.99 with CodeGalaxy Z Fold 8 inner screen corners reported with excessive giveHuawei unveils Mate XT 2 tri-fold with privacy-screen edition as Apple prepares folding iPhoneWall Street Analysts Recommend Nvidia, Uber and Marvell for Long-Term GrowthLenovo, AMD and Huawei Put Local Large-Model AI at Center of New Devices
AI Chips & Compute

CoreWeave Leads GPU Neocloud Ranking as Only Platinum-Rated Provider

CoreWeave tops GPU neocloud ranking as only Platinum provider, while Nebius, Lambda, Crusoe, and Groq differentiate on pricing, power, and hardware.

AI News Feed · 2026-08-24 15:09

CoreWeave reported second-quarter 2026 revenue of $2.575 billion, up 112% year over year, with a net loss of $626 million. Its published price for an NVIDIA HGX H100 is $6.16 per GPU-hour, roughly 10–15% higher than comparable rates from Nebius, Crusoe, and Lambda on managed clusters, according to SemiAnalysis. The company disclosed a backlog of about $104 billion, which includes deals with Meta, Anthropic, and OpenAI, and raised its 2026 capital expenditure guidance to $35–39 billion.

Nebius, a public company filing with the SEC as a foreign private issuer, reported AI cloud annual recurring revenue of $3.0 billion in Q2 2026, up 454% year over year. It is the only provider in the group publishing on-demand pricing for the B300 and lists the lowest on-demand H100 rate at $3.85 per GPU-hour. Nebius also priced $5.0 billion in convertible notes on August 19, 2026.

Lambda, which is private and preparing for an IPO in the second half of 2026, publishes the lowest on-demand B200 rate at $6.69 per GPU-hour but offers no spot tier. Its CEO, Michel Combes, took the role in May 2026, according to the article.

Crusoe, also private, is the only provider here that lists AMD MI300X and MI355X on its rate card. It has 4.9 GW of contracted power and a pipeline of more than 40 GW. Crusoe's Abilene site is being energized, and it has secured a 1.2 GW deal with Oracle and OpenAI, plus a 900 MW campus for Microsoft.

Groq, which licensed its LPU technology to NVIDIA in December 2025, refocused entirely on inference-cloud infrastructure. It now operates 13 data centers with 54 MW of power, expanding to over 200 MW by 2027. In August 2026, Groq raised $350 million in a Series A at a $3.5 billion valuation, with NVIDIA participating.

The analysis also covers hardware roadmaps. CoreWeave reported first bring-up and validation of Vera Rubin NVL72 systems in Q2 2026, while Nebius has received Vera Rubin NVL72 units for validation. Lambda is a launch partner for Vera CPU and Quantum-X800 CPO, and Crusoe supports Vera CPU launch with MI355X via sales. Groq became an NVIDIA Cloud Partner on August 12, 2026, and plans to add NVIDIA GPU capacity alongside its LPU inference cloud.

Contract structures vary. CoreWeave offers reserved discounts up to 60% off with free egress; Nebius offers commit discounts up to 35% off and free network egress; Lambda and Crusoe are sales-quoted with free egress and no spot tier. Groq uses per-token pricing with sales-led clusters.

The comparison draws on public pricing pages, quarterly reports, and the SemiAnalysis ClusterMAX 2.0 rating. As of the analysis, CoreWeave remains the only Platinum-rated provider in ClusterMAX, receiving the rating twice.