Super Micro Computer Inc. (SMCI) reported its second-quarter fiscal 2026 earnings after market close on February 3, delivering a massive revenue beat but revealing significant pressure on profitability. The AI server manufacturer reported revenue of $12.68 billion, far surpassing the Wall Street consensus estimate of $10.42 billion and more than doubling its Q1 2026 revenue of $5.0 billion. This marks the company's highest quarterly revenue to date.
On the bottom line, net income for the quarter was $401 million, up from $168 million last quarter. Diluted earnings per share (EPS) came in at $0.60 on a GAAP basis, beating the analyst forecast of $0.49. On a non-GAAP basis, EPS was $0.69, compared to $0.59 in the same quarter last year.
However, the stellar revenue was overshadowed by a sharp contraction in gross margins. The GAAP gross margin fell to 6.3%, a significant drop from 9.3% in Q1 and 11.8% in Q2 of the prior year. Non-GAAP gross margin was 6.4%, down from 11.9% a year ago. This margin pressure is attributed to intense competition in the AI server space and rising input costs, as highlighted by analysts like Goldman Sachs' Mike Ng, who maintains a Sell rating on the stock.
The company's balance sheet nearly doubled in six months, with total assets reaching $28.0 billion, driven by a surge in accounts receivable to $11.0 billion and inventories to $10.6 billion. Conversely, cash flow was negative, with the company using $941.4 million in net cash from operations over the six-month period due to heavy investments in working capital.
Looking ahead, SMCI provided guidance for Q3 fiscal 2026, expecting revenue of at least $12.3 billion and GAAP EPS of $0.52. For the full fiscal year 2026, the company anticipates sales of at least $40.0 billion. CEO Charles Liang credited the company's rapid growth to strong demand for its AI server and storage systems and its Data Center Building Block Solutions.
The earnings report comes amid a lingering investigation by the U.S. Justice Department into the company's accounting practices, a concern previously highlighted by short-seller Hindenburg Research.
