TEM ·深度研究报告 · In-Depth Research Report

分享
TEM ·深度研究报告 · In-Depth Research Report

Q4营收$3.672亿同比增长83%、FY2025全年营收$12.7亿同比增长83.4%、诊断业务营收$2.669亿同比增长121.6%、数据与应用业务营收$1.004亿创历史纪录同比增长25.1%其中Insights数据授权增长69.5%、总剩余合同价值突破$11亿、净营收留存率126%、现金及有价证券$7.597亿、全年调整后EBITDA亏损$740万(较上年同期改善$9730万)、2026年营收指引$15.9亿同比增长约25%并首次给出正调整后EBITDA约$6500万承诺——Cathie Wood ARK于2026年2月25日单日买入21.3万股、摩根士丹利目标价今日下调至$70引发市场情绪压制、股价从上市后高点大幅回落至约$50至$55区间:全球最大多模态临床AI数据平台,诊断业务是现金流基本盘,Insights数据授权是高利润率第二曲线,数字病理Paige Predict是第三引擎,EBITDA转正倒计时已经启动,临床AI决策基础设施最深数据护城河的错杀建仓窗口

Q4 revenue of $367.2 million grew 83% year-over-year; FY2025 full-year revenue of $1.27 billion grew 83.4% year-over-year; diagnostics revenue of $266.9 million grew 121.6% year-over-year; data and applications revenue of $100.4 million set an all-time record, growing 25.1% year-over-year, with Insights data licensing growing 69.5%; total remaining contract value exceeded $1.1 billion; net revenue retention rate was 126%; cash and marketable securities were $759.7 million; full-year adjusted EBITDA loss narrowed to $7.4 million, an improvement of $97.3 million year-over-year; 2026 revenue guidance of approximately $1.59 billion implies approximately 25% year-over-year growth, with the first-ever commitment to positive adjusted EBITDA of approximately $65 million — Cathie Wood's ARK purchased 213,000 shares in a single day on February 25, 2026; Morgan Stanley lowered its target price to $70 today, creating sentiment pressure; shares have pulled back sharply from post-IPO highs to the approximately $50 to $55 range: the world's largest multimodal clinical AI data platform, with diagnostics as the cash flow foundation, Insights data licensing as the high-margin second growth curve, digital pathology Paige Predict as the third engine, and the EBITDA breakeven countdown now officially underway — a mispriced entry window into the deepest data moat in clinical AI decision infrastructure

2016年,Eric Lefkofsky的妻子被诊断为乳腺癌。他发现医生在用直觉和经验做决定,而不是数据。现代医疗系统的数据基础设施几乎是原始的——电子病历是孤岛,基因组数据是碎片,影像数据是黑盒,临床试验数据是密室。没有人能把这些东西系统性地整合在一起,让AI看到一个完整的病人。

In 2016, Eric Lefkofsky's wife was diagnosed with breast cancer. He discovered that doctors were making decisions based on intuition and experience rather than data. The data infrastructure of modern healthcare systems was nearly primitive — electronic medical records were siloed, genomic data was fragmented, imaging data was a black box, and clinical trial data was locked away. No one had been able to systematically integrate all of these elements in a way that allowed AI to see a complete picture of a patient.

十年后,Tempus AI用全球最大的多模态临床数据库之一,正在完成医疗决策智能化基础设施的建设:超过200万患者的基因组数据、病理影像、电子病历、治疗记录,被结构化地整合进一个可以被AI系统性分析的平台。合作医院网络覆盖全美超过六成的肿瘤科室。与阿斯利康、强生、诺华的战略合作相继落地,验证了Tempus数据资产在全球头部药企端的商业价值。

Ten years later, Tempus AI is completing the construction of an intelligent infrastructure for medical decision-making using one of the world's largest multimodal clinical databases: genomic data, pathological imaging, electronic medical records, and treatment records for more than two million patients have been structurally integrated into a platform that AI can systematically analyze. The partner hospital network covers more than 60% of oncology departments across the United States. Strategic partnerships with AstraZeneca, Johnson & Johnson, and Novartis have been established one after another, validating the commercial value of Tempus's data assets among the world's leading pharmaceutical companies.

2026年2月24日,Tempus AI公布了FY2025全年业绩。数字清晰而令人印象深刻:Q4营收$3.672亿,同比增长83%;全年营收$12.7亿,同比增长83.4%;Q4诊断业务营收$2.669亿,同比增长121.6%,肿瘤基因组检测量同比增长29%,遗传检测量增长23%;Q4数据与应用业务营收$1.004亿创历史单季纪录,同比增长25.1%;Insights数据授权同比增长69.5%;总剩余合同价值突破$11亿;净营收留存率126%,代表现有客户在下一年度贡献的营收比当年增长26%;全年毛利润$7.979亿,同比增长109.4%;全年调整后EBITDA亏损$740万,较上年同期改善$9730万;现金及有价证券$7.597亿。

On February 24, 2026, Tempus AI reported its FY2025 full-year results. The numbers were clear and impressive: Q4 revenue of $367.2 million grew 83% year-over-year; full-year revenue of $1.27 billion grew 83.4% year-over-year; Q4 diagnostics revenue of $266.9 million grew 121.6% year-over-year, with oncology genomic test volume growing 29% year-over-year and hereditary testing volume growing 23%; Q4 data and applications revenue of $100.4 million set an all-time single-quarter record, growing 25.1% year-over-year; Insights data licensing grew 69.5% year-over-year; total remaining contract value exceeded $1.1 billion; net revenue retention rate was 126%, meaning existing customers contributed 26% more revenue in the following year compared to the current year; full-year gross profit was $797.9 million, growing 109.4% year-over-year; full-year adjusted EBITDA loss narrowed to $7.4 million, improving by $97.3 million year-over-year; and cash and marketable securities were $759.7 million.

同时,管理层给出了2026年全年指引:营收约$15.9亿,同比增长约25%;调整后EBITDA约正$6500万——这是Tempus AI历史上首次给出正EBITDA的全年指引,是公司从高增速烧钱模式切换至盈利模式的历史性时刻。

At the same time, management provided 2026 full-year guidance: revenue of approximately $1.59 billion, implying approximately 25% year-over-year growth; adjusted EBITDA of approximately positive $65 million — this is the first time in Tempus AI's history that it has provided a full-year guidance for positive EBITDA, marking a historic moment in the company's transition from a high-growth cash-burning model to a profitable one.

今天(2026年3月3日)的股价:约$50至$55。摩根士丹利今日将目标价从约$85下调至约$70,成为短期情绪压制的主要来源。而与此同时,Cathie Wood的ARK在2026年2月25日财报后单日买入21.3万股——两个截然不同的机构行动,正在同一家公司上演,制造了今天的价格。

Today's stock price as of March 3, 2026 is approximately $50 to $55. Morgan Stanley lowered its target price from approximately $85 to approximately $70 today, becoming the primary source of short-term sentiment pressure. At the same time, Cathie Wood's ARK purchased 213,000 shares in a single day following the earnings release on February 25, 2026 — two completely different institutional actions playing out in the same company, creating today's price.

本报告将深入拆解Tempus AI的四大核心投资逻辑、关键催化剂时间节点、华尔街多空双方最新目标价对比,以及我们独立得出的12至24个月目标价区间——数字或许会让你意外。

This report will break down Tempus AI's four core investment pillars, key catalyst timelines, the latest Wall Street bull and bear target price comparisons, and our independently derived 12- to 24-month target price range — and the numbers may surprise you.

阅读更多

周期转折交易系统 . Cycle Trading System

周期转折交易系统 . Cycle Trading System

【主图 · Main Chart】 当大多数人还在一根一根K线里猜方向,Cycle Trading System已经先把趋势、节奏和拐点切分出来了。 While most traders are still guessing direction candle by candle, the Cycle Trading System has already mapped out the trend, rhythm, and turning points for you. 市场里真正难的,从来不是看见涨跌,而是分清楚眼前这段上涨到底只是反弹,还是一段能继续推进的趋势;眼前这段下跌到底只是洗盘,还是结构已经开始转弱。多数散户之所以反复追高杀低,不是因为不努力,而是因为他们看到的只是价格本身,却看不到价格背后的节奏切换、趋势延续和转折质量。 The real difficulty in trading

lock-1 By Wealth Club
错失的代价vs长期机遇 / The Cost of Missing Out vs Long-term Opportunity

错失的代价vs长期机遇 / The Cost of Missing Out vs Long-term Opportunity

杠杆交易员的抛售压力 本周市场出现一轮调整,背后的主要推手是杠杆交易员的风险规避行为。就业报告强于预期,市场对美联储可能加息的预期推高了收益率。当利率上升时,借入大量资金进行交易的机构面临借贷成本上升,他们通常会迅速减仓以降低杠杆敞口。科技股首当其冲,因为该板块流动性最强,也拥有最多的衍生品头寸——包括个股期货、基金持有量和期权。不过,单日波动并不构成趋势,市场短期调整属于正常现象。 Leverage-fueled selloff pressure This week's market correction is primarily driven by risk-averse deleveraging from leveraged traders. A stronger-than-expected employment report pushed yield expectations higher, with markets pricing in potential Fed rate hikes.

lock-1 By Wealth Club
量化趋势交易系统 · QuantFlow Trading System

量化趋势交易系统 · QuantFlow Trading System

【Main Chart】 市场每天都在运动,但真正值得参与的机会,永远只在趋势明确的那一刻。 Markets move every day — but the opportunities truly worth acting on only appear at the moment a trend becomes clear. 大多数人错过最好的买点,不是因为反应慢,而是因为他们无法在噪音中识别真正的趋势转变。等到方向人人都看见了,最好的入场时机已经过去。QuantFlow Trading System · 量化趋势交易系统由财富俱乐部量化分析团队打造,专门解决这一个问题——通过多维度量化模型,提前捕捉趋势的启动与终结,让你在关键转折点做出清晰决定。 Most people miss the best entry points not because they react slowly,

lock-1 By Wealth Club
NVIDIA与ASIC芯片竞争格局重塑:Computex 2026的市场启示 NVIDIA and ASIC Competition Redefined: Market Insights from Computex 2026

NVIDIA与ASIC芯片竞争格局重塑:Computex 2026的市场启示 NVIDIA and ASIC Competition Redefined: Market Insights from Computex 2026

近期市场普遍担忧ASIC(定制芯片,如Google TPU)会对NVIDIA的GPU垄断地位造成50/50的分割。然而,来自NVIDIA内部的深入观点表明这一担忧被过度放大。虽然Google TPU在成本上具有竞争力(约1500-2000 token/秒),但NVIDIA H100在优化配置下可达2000-3500 token/秒,性能优势达25%-133%。更关键的是,市场分割更可能呈现80/20或70/30格局,大幅倾斜于NVIDIA。TPU的优势仅限于简单推理任务(如图像生成),而复杂、高吞吐工作负载仍然依赖NVIDIA芯片。Google为推动TPU采用已向客户提供显著折扣,这本身反映了市场权力失衡。 Recent market concerns over ASIC chips (custom silicon like Google TPU) creating a 50/50 split in the GPU market

By Wealth Club