General Compute secures $100 million Series A as neocloud market faces turmoil

Market backdrop and General Compute’s raise

CoreWeave, a publicly traded neocloud provider, has seen its share price slump to a twelve‑month low while Elon Musk has warned that more than two hundred neoclouds could face “guppy‑level death and consolidation.”

Musk’s comments appear aimed at driving down GPU prices for the computing capacity he intends to purchase, a strategy that has added pressure to the sector.

Finn Puklowski, co‑founder and CEO of General Compute, remains unfazed and says he is bullish on CoreWeave’s chances despite the negative sentiment.

Against that backdrop, General Compute announced a $100 million Series A round, positioning the startup as a counterpoint to the prevailing “neocloud disaster” narrative.

Strategic focus on inference chips

Puklowski told TechCrunch that fears of a chip‑led collapse are “massively overstated” and that he receives weekly requests for TPUs, GPUs and other processors.

The company has built its cloud around specialized inference chips, signing a $400 million purchase agreement with SambaNova—the first debt financing for such chips—and adding Positron and Cerebras to its supplier roster.

General Compute’s approach is to “zig where others zag,” concentrating on chips built for inference while most competitors chase general‑purpose GPUs.

Instead of constructing massive data‑center campuses, the startup rents space in existing facilities and repurposes former bitcoin‑mining sites to host AI models, a tactic that reduces upfront infrastructure costs.

This model means the firm can plug new chips into already available power sources rather than waiting for a purpose‑built campus to become operational.

Funding, valuation and growth outlook

A $100 million customer contract delivering $100 million in annual recurring revenue helped secure the Series A, which was led by Blackbird Ventures and Abstract Ventures.

Following a May seed round that valued the firm at $60 million, the new financing lifts its post‑money valuation to $800 million.

Puklowski projects the company will deploy 75 megawatts of compute capacity by 2027 and believes access to 200 megawatts of power could be within reach, though he admits that “capital is going to be a full‑time job for me, both on debt and equity.”

General Compute supplies bare‑metal servers—raw hardware without a software layer—to cloud companies that specialize in high‑speed inference, enabling those partners to bring developer‑focused services to market quickly.

James Palmer, a principal at Blackbird who co‑led the round, highlighted the firm’s relevance by referencing Blackbird’s investment in Baseten and the growing demand for open‑weight inference and self‑hosted intelligence.

Palmer added that “Cerebras, Positron, SambaNova, all need distribution partners who can take them to scale, who can navigate the complexity of the physical world.”

Rodrigo Liang, CEO and co‑founder of SambaNova, echoed this view, stating that “everyone in this market is selling speed, the real question is whether you can make money doing it.”

Puklowski also noted that Anthropic’s bullish S‑1 filing is “settling fears for today,” suggesting that broader market confidence is improving.

While General Compute does not aim to become a top‑tier neocloud or challenge hyperscalers immediately, its flexible infrastructure and focus on inference economics position it to serve a segment of the AI compute market that is currently underserved.

Why This Matters: General Compute’s funding validates a business model that sidesteps the massive capital outlays of traditional neoclouds, offering a potentially more sustainable path for AI inference providers.