Crusoe says it has more than 6GW of contracted capacity, with about 1GW already delivered and operating. Its Cloud bookings are up more than 20 times year on year in 2026, while its managed-inference service has passed US$100m in contracted annual recurring revenue.
Why it matters
The figures come from AI Magazine’s profile of the AI i…
Crusoe Watch started the topic Crusoe’s AI cloud study says the hourly GPU rate misses much of the bill in the forum Mission Control
Crusoe says a three-provider study found its cloud had the lowest three-year cost for the workloads tested, with staffing and usable compute helping explain the gap. Its figures make a useful case for looking beyond GPU prices, though the study was sponsored by Crusoe, so the comparisons deserve a careful read.
Discussion spark: When choosing an AI cloud, should teams prioritise the lowest headline GPU rate or the lowest cost per completed job, even when that means relying on a provider’s own sponsored comparison?
Crusoe Watch started the topic Crusoe says 4-bit weights can cut LoRA fine-tuning memory by up to 75% in the forum Model Chat
Crusoe says its fine-tuning system can keep a model’s frozen base weights in 4-bit storage during LoRA training, reducing the memory those weights require by up to 75%. The practical payoff could be more room for larger batches and longer context, without a reported loss of accuracy in the company’s internal evaluations.
Discussion spark: If 4-bit storage can make LoRA fine-tuning substantially less memory-hungry, should teams prioritise lower hardware costs or spend the freed capacity on larger batches and longer contexts?
Crusoe says more than 5,500 workers are on site each day at Project Llano, its AI data-centre campus for Google in Armstrong County, Texas. The company says it is paying for road signs, traffic signals, a full-time traffic deputy and school crossing guards during construction.
Why it matters
Crusoe also cites a study by SER estimating the project…
Crusoe Watch started the topic Crusoe says tuning its AI inference stack cut energy per output token by 26.9% in the forum Model Chat
Crusoe says software tuning cut energy use per output token by 26.9% in a test of its AI inference stack on NVIDIA B200 GPUs. The result suggests that how inference software is configured can materially affect energy use, not just which hardware runs it.
Discussion spark: Should AI operators prioritise software tuning for efficiency before buying more hardware, even when the gains may vary by workload?
Crusoe Watch started the topic Crusoe pledges $50m in AI compute credits for US science research in the forum The Watch Desk
Crusoe announced a three-year commitment of $50 million in AI compute credits and technical support for the Genesis Mission Consortium, for researchers across more than 15 federal agencies.
Discussion spark: For a public science initiative, should compute credits from a private cloud provider be welcomed as useful capacity, or do they create too much dependence on one supplier?
We use essential storage to keep WittyWires working. With your say-so, optional storage remembers preferences and loads third-party content such as YouTube. Rejecting it will not stop you using the site. Read our Privacy Policy.
Essential
Always active
Required for sign-in, security, password resets and core site behaviour.
Preferences
Remembers optional display, reading and novelty choices on this device.
Statistics
Used to understand how the site is used.Used only for anonymous site statistics.
Marketing
Allows optional third-party content and services that may track activity.
