Speed to power beats cheap power for AI data centres

A Carnegie Mellon and Berkeley preprint prices the months a data centre stands idle while it waits for electricity, and finds that off-grid solar with batteries wins in 85% of the continental US even though it is rarely the cheapest source per kilowatt-hour. Anthropic's search for a gas supply chief shows the other bet, and the model's limits show where it could fail.

14 November 2026 · 7 min · AI-generated

CoreWeave Prices Short GPU Rentals Far Above Long Contracts

CoreWeave says it raised GPU prices about 25% in July and 10% later, and now sells short-term capacity at more than $40 million per megawatt a year against $20 to $25 million on multi-year deals. The premium pays the company, but lenders prefer the cheaper long contracts, and its bonds show their doubt.

12 November 2026 · 6 min · AI-generated

AI's suppliers fell faster than their biggest spenders

Since 6 October the firms that build and equip AI data centres have lost a median 23% of their share price, against 5% for the Big Tech companies paying them, even as backlogs hit records. Two weeks of evidence points to late power, costlier financing and local opposition as the pressure points.

24 October 2026 · 7 min · AI-generated

Microsoft leaves Nscale's Essex site as the grid runs out

Microsoft has reportedly pulled out of Nscale's flagship UK AI data centre because the power will not arrive until the 2030s. A separate audit of US moratoriums suggests that wires, more than vetoes, decide how fast AI capacity gets built.

19 October 2026 · 7 min · AI-generated

PolyStrata runs large MoE models on gaming GPUs

The open-source runner keeps selected model experts in system memory and moves them to a consumer GPU as needed. Its one-click package exposes local OpenAI- and Anthropic-compatible APIs.

12 October 2026 · 3 min · AI-generated

rGPU turns a remote card into a PyTorch device

The Apache-licensed project keeps Python and model code on a client machine while tensors and CUDA operations run on a remote NVIDIA GPU.

8 October 2026 · 2 min · AI-generated

OpenTPU runs local models on a $300 FPGA card

The Apache-licensed project publishes its accelerator design, compiler, simulator and host tools, with measured throughput showing a small, memory-bound system rather than a GPU replacement.

7 October 2026 · 2 min · AI-generated

vLLM shows when split serving helps and hurts

A practical guide measures the trade-off in separating prompt processing from token generation. Tail latency improves under load, but moving the model's working memory delays the first token.

6 October 2026 · 3 min · AI-generated

Strata fork revives an IBM AI server for local LLMs

A hardware-specific fork runs Qwen3.8-Flash-Next across two POWER9 processors and four V100 GPUs, showing what model-aware optimisation can recover from a 2018 system.

5 October 2026 · 3 min · AI-generated

Bull doubles French supercomputer output

Bull expanded its Angers plant from six to 12 supercomputer racks a month as Europe builds more sovereign AI capacity.

2 October 2026 · 3 min · AI-generated

HPE lands $1.2 billion Vultr AI rack order

Vultr ordered AMD Helios racks and HPE networking for US data centres. It is HPE's first order for the system and a large commercial test of an Ethernet-based AI stack.

1 October 2026 · 4 min · AI-generated

Dutch registry misses nearly half of large data centres

A Europe-wide investigation found major gaps between mandatory data-centre reporting and public records. In the Netherlands, the registry lists 104 of 186 qualifying commercial sites.

1 October 2026 · 4 min · AI-generated