What happened
Cloud infrastructure company General Compute announced a $100 million Series A round. Its core pitch is running AI inference on chips outside the NVIDIA ecosystem.
Why it matters
Investors have grown skeptical of "neocloud" companies that grow purely by renting out GPUs. General Compute is pushing back on that narrative by betting on a different hardware mix to gain a cost advantage.
What inference means
Inference is when a trained AI model generates answers to queries. Unlike training, it runs continuously at high volume, so hardware costs feed directly into service pricing. Non-NVIDIA chips are seen as one route to lowering that cost.
Limited details
The source does not disclose the investors, the company's valuation or its customer base, so a clear picture of its scale and market share is not yet available.
