广告
加载中

英伟达推算力换营收方案 扶持AI初创企业

亿邦AI 2026-07-03 14:29
亿邦AI 2026/07/03 14:29

邦小白快读

EN
全文速览

这篇文章核心介绍了英伟达最新推出的针对AI初创企业的合作项目,同时披露了英伟达最新的融资动态,核心干货如下:

1. 项目核心规则:英伟达为参与项目的AI初创企业对接由自家芯片支撑的全栈计算资源,以算力额度换取这些企业未来产品、云业务一定比例的营收分成,参与项目的所有相关企业都要按约定给英伟达分营收。

2. 项目现有算力储备:首批有两家合作方提供算力支撑,澳大利亚Sharon AI部署最多4万台英伟达GPU,新加坡Firmus Technologies在印尼建360兆瓦数据中心,可容纳最多17万台英伟达GPU。

3. 配套保障与融资动态:英伟达给参与的云服务商提供财务担保,闲置GPU由英伟达回租承担成本,另外英伟达刚计划募资至少200亿美元用于企业运营与再融资。

英伟达此次推出的新合作模式,对AI领域品牌布局、市场拓展有较高的参考价值,核心干货如下:

1. 品牌业务布局参考:英伟达打破传统直接卖芯片给头部大客户的模式,通过算力换营收的方式绑定大量新兴AI初创企业,还配套风险兜底措施降低合作方门槛,能有效分散业务风险,降低自身对亚马逊、微软等头部大客户的依赖,应对头部客户自研硬件的竞争冲击。

2. 行业合作趋势:当前AI行业普遍存在初创企业算力不足、GPU供给不稳成本高的问题,业内已经出现不少芯片厂商和AI企业以营收、股权分成合作缓解压力的案例,说明这种上下游风险共担、资源绑定的模式会成为AI行业新的合作趋势。

3. 新兴市场机会:品牌商可以参考该模式,提前布局绑定新兴AI初创品牌,抓住AI行业高速增长的红利,开拓新的营收增长点。

对于AI领域创业卖家、相关从业者来说,这次英伟达的新项目带来了明确的机会与需要注意的风险,核心内容如下:

1. 明确的创业机会:当前AI初创企业最大的门槛就是算力不足,前期采购GPU需要投入大量资金,很多项目卡在算力环节无法推进,该项目不需要初创企业前期大额投入,只需要出让未来一定比例营收就能获得稳定算力,大大降低了AI创业的门槛,中小创业者可以抓住该机会对接资源。

2. 风险提示:参与项目需要长期向英伟达出让营收分成,会分流自身项目的利润,创业者在合作前需要做好成本收益测算,避免后续出现入不敷出的问题。

3. 行业风向:英伟达目前正在调整客户结构,逐步降低对头部科技巨头的依赖,会向中小AI创业项目倾斜更多资源,这对中小卖家来说是难得的发展窗口期。

对于AI硬件生产、数据中心基建相关工厂来说,本次英伟达的新项目带来了明确的商业机会和发展启示,核心内容如下:

1. 短期商业机会:从项目现有规划来看,首批两家合作方就要部署合计超过21万台英伟达GPU,还要建设规模达360兆瓦的大型数据中心,这给上游GPU配套生产、基建工程、机电配套等工厂带来了大量的订单需求,相关工厂可以积极对接该项目的配套需求,获取新的营收增长。

2. 长期行业需求趋势:当前AI行业算力缺口是普遍问题,越来越多企业会采用这种算力扩张模式,未来对GPU、数据中心基础设施的需求会长期维持高位,相关工厂可以提前布局对应产能,匹配行业增长需求。

3. 转型启示:AI产业的高速扩张带动全行业数字化转型,相关工厂也可以抓住产业风口,推进自身数字化、智能化升级,更好地对接AI产业的定制化生产需求。

对于AI算力服务商、云服务商等相关服务商来说,本文披露了行业最新的发展趋势与可参考的解决方案,核心干货如下:

1. 当前核心客户痛点:下游AI初创企业、中小AI开发者的核心痛点是算力供给不稳定、前期投入成本过高,算力缺口长期存在,这是服务商拓展业务的核心切入点。

2. 可参考的合作解决方案:英伟达推出的“算力换营收”模式,配套了完善的风险保障机制,给参与合作的云服务商提供财务担保,闲置的GPU由英伟达回租,承担未用算力的成本,还能同时满足GPU和数据中心的融资需求,这种风险共担的模式很好地解决了服务商扩产的后顾之忧,值得服务商参考借鉴。

3. 行业发展机会:目前英伟达正在调整客户结构,降低对头部科技巨头的依赖,加大对中小AI项目的资源投入,服务商可以积极对接英伟达的该项目,拓展自身的客户群体,获得更多业务增长空间。

对于AI算力平台、科技平台等平台商来说,本次英伟达的新项目带来了运营管理、生态建设的诸多参考,核心内容如下:

1. 平台用户核心需求:当前AI领域的参与者,尤其是中小初创参与者,最核心的需求就是稳定、低成本的算力供给,以及配套的风险保障与融资支持,平台生态建设需要围绕这个核心痛点搭建服务体系。

2. 可参考的平台运营玩法:英伟达推出的算力换营收分成模式,通过资源换长期收益的方式既绑定了合作方,又解决了初创企业的现金流痛点,还配套了风险兜底机制,给合作服务商提供财务担保,回租闲置算力降低合作方风险,这种风险共担、利益共享的模式非常适合平台搭建生态,可以借鉴用来吸引更多中小经营者入驻。

3. 风向提示:当前头部云厂商都在布局自研AI硬件,上游芯片厂商开始分散客户结构,平台可以抓住这个产业调整的窗口期,加大AI领域招商力度,吸纳更多中小AI企业和算力服务商,完善平台的AI生态布局。

对于AI产业研究者来说,本文披露了AI芯片领域最新的产业动向与商业模式创新,核心研究价值干货如下:

1. 全新的商业模式创新:英伟达打破了传统芯片厂商直接卖芯片获取营收的模式,推出“算力换营收”的全新合作模式,面向AI初创企业输出算力,换取未来长期营收分成,还配套了财务担保机制,解决合作方的闲置算力风险,同时可以覆盖GPU和数据中心的融资需求,是AI产业上游非常有创新性的商业模式探索。

2. 产业结构变动新动向:目前英伟达的核心客户是亚马逊、微软、谷歌等头部科技巨头,而这些巨头都在推进自研AI硬件布局,降低对外部芯片的依赖,英伟达推出该项目就是为了降低对头部大客户的依赖,拓展中小客户群体,反映出AI芯片产业供应链正在发生结构性变化。

3. 产业共性问题的反映:该模式的出现也印证了当前AI行业初创企业普遍面临算力稀缺、GPU供给不稳定、融资难、现金流压力大的共性问题,该模式为解决行业流动性压力提供了新的解决路径,值得深入研究。

返回默认

声明:快读内容全程由AI生成,请注意甄别信息。如您发现问题,请发送邮件至 run@ebrun.com 。

我是 品牌商 卖家 工厂 服务商 平台商 研究者 帮我再读一遍。

Quick Summary

This article outlines NVIDIA's newly launched partnership program for AI startups and discloses the company's latest fundraising update. Key takeaways are as follows:

1. Core program rules: NVIDIA provides participating AI startups with access to full-stack computing resources powered by its chips, in exchange for a percentage of future revenue from the startups' products and cloud services. All participating companies are required to share revenue with NVIDIA per the agreement.

2. Initial computing capacity: Two founding partners are supporting the program's first batch of capacity. Australia-based Sharon AI will deploy up to 40,000 NVIDIA GPUs, while Singapore's Firmus Technologies is building a 360-megawatt data center in Indonesia that can accommodate up to 170,000 NVIDIA GPUs.

3. Supporting safeguards and fundraising: NVIDIA provides financial guarantees to participating cloud service providers and covers costs for idle GPUs through a buy-back lease arrangement. Additionally, NVIDIA recently announced plans to raise at least $20 billion for corporate operations and refinancing.

NVIDIA's new partnership model offers valuable insights for brand positioning and market expansion in the AI industry. Key takeaways are as follows:

1. Implications for brand strategy: NVIDIA has moved beyond its traditional model of selling chips directly to large enterprise clients. By tying up with a large base of emerging AI startups through a "compute capacity for revenue" model, paired with risk-mitigation measures that lower the barrier to entry for partners, the company effectively diversifies business risk, reduces its reliance on major clients like Amazon and Microsoft, and buffers against competition from large clients' in-house hardware development efforts.

2. Emerging industry collaboration trend: AI startups broadly face challenges of insufficient compute capacity, unstable GPU supply and high costs. A growing number of chipmakers and AI companies have already adopted revenue or equity-sharing partnerships to alleviate these pressures, indicating that this model of upstream-downstream risk-sharing and resource binding will become a new mainstream collaboration trend in the AI sector.

3. Opportunities for emerging market positioning: Brands can adopt this model to proactively partner with emerging AI startups early, capture growth dividends from the fast-expanding AI industry, and unlock new revenue streams.

For AI startup founders and industry practitioners, NVIDIA's new program brings clear opportunities and notable risks. Key takeaways are as follows:

1. Clear startup opportunities: The biggest barrier for AI startups today is limited compute capacity, as upfront GPU procurement requires massive capital investment that leaves many projects stalled. This program eliminates the need for large upfront investment from startups, which gain access to stable compute capacity in exchange for a share of future revenue, significantly lowering the barrier to entry for AI entrepreneurship. Small and medium-sized founders can leverage this opportunity to access critical resources.

2. Risk warning: Participating startups are required to cede a share of revenue to NVIDIA over the long term, which reduces project profit margins. Founders should conduct thorough cost-benefit analysis before entering the partnership to avoid eventual insolvency.

3. Industry trend indication: NVIDIA is currently restructuring its customer base to gradually reduce reliance on large tech giants, and will allocate more resources to small and medium-sized AI projects. This creates a rare window of opportunity for smaller industry players to grow.

For factories engaged in AI hardware manufacturing and data center infrastructure, NVIDIA's new program opens clear business opportunities and strategic insights. Key takeaways are as follows:

1. Immediate near-term business opportunities: Per the program's initial plans, the two founding partners will deploy more than 210,000 NVIDIA GPUs in total and build a 360-megawatt large-scale data center. This creates massive order demand for upstream GPU component manufacturing, infrastructure construction, electrical and mechanical supporting facilities, and other related manufacturing sectors. Relevant factories can proactively pursue supporting contracts for the program to drive new revenue growth.

2. Long-term industry demand trend: The compute capacity shortage is a widespread problem across the AI industry, and more companies will adopt this capacity expansion model. This means demand for GPUs and data center infrastructure will remain at high levels over the long term. Relevant manufacturers can proactively expand targeted production capacity to align with industry growth.

3. Implications for transformation: The rapid expansion of the AI industry is driving digital transformation across all sectors. Related factories can capitalize on this industry boom to advance their own digital and intelligent upgrading, to better meet the customized manufacturing needs of the AI sector.

For AI compute providers, cloud service providers and other related service providers, this article outlines the latest industry trends and actionable solution frameworks. Key takeaways are as follows:

1. Core pain points of current clients: Downstream AI startups and small and medium-sized AI developers broadly face unstable compute supply, high upfront costs, and persistent capacity gaps, which represent a key entry point for service providers looking to expand their business.

2. A replicable collaborative solution: NVIDIA's "compute capacity for revenue" model includes a comprehensive risk-mitigation framework: it provides financial guarantees to participating cloud providers, leases back idle GPUs to cover costs for unused capacity, and meets financing needs for both GPU procurement and data center development. This risk-sharing model effectively removes barriers to capacity expansion for service providers, making it a valuable framework to replicate.

3. New growth opportunities: NVIDIA is restructuring its customer base to reduce reliance on large tech giants and increase resource allocation to small and medium-sized AI projects. Service providers can partner with NVIDIA on this program to expand their customer base and unlock new growth opportunities.

For AI compute platforms and tech marketplaces, NVIDIA's new program offers multiple insights for operation management and ecosystem building. Key takeaways are as follows:

1. Core user demand: The most critical demand for AI industry participants, especially small and medium-sized startups, is stable, low-cost compute access, paired with risk mitigation and financing support. Platform ecosystem building should center its service framework around solving this core pain point.

2. A replicable platform operation model: NVIDIA's revenue-sharing compute model binds partners to the ecosystem through resource exchange for long-term returns, solves the cash flow crunch facing startups, and includes built-in risk mitigation (financial guarantees for partner service providers and leasebacks for idle capacity to reduce partner risk). This risk-sharing, profit-sharing model is ideal for platform ecosystem building, and can be adapted to attract more small and medium-sized operators to the platform.

3. Strategic industry indication: Leading cloud providers are all developing their own in-house AI hardware, while upstream chipmakers are diversifying their customer bases. Platforms can capitalize on this industry restructuring window to step up recruitment efforts in the AI sector, onboard more small and medium-sized AI companies and compute service providers, and完善 their AI ecosystem布局。

For AI industry researchers, this article discloses the latest industry developments and business model innovation in the AI chip sector. Key research takeaways are as follows:

1. A groundbreaking new business model innovation: NVIDIA has broken away from the traditional chipmaker model of generating revenue via direct chip sales, and launched an entirely new "compute capacity for revenue" partnership model. It provides compute capacity to AI startups in exchange for a share of long-term future revenue, and includes supporting financial guarantees to resolve idle capacity risk for partners while meeting financing needs for both GPUs and data centers. This represents a highly innovative business model experiment for upstream players in the AI industry.

2. New dynamics of industrial structural change: NVIDIA's core clients are currently large tech giants including Amazon, Microsoft and Google, all of which are developing in-house AI hardware to reduce reliance on external chip suppliers. NVIDIA's new program is designed to cut its dependence on large anchor clients and expand its small and medium-sized customer base, reflecting the ongoing structural shift in the AI chip industry supply chain.

3. Reflection of common industry-wide challenges: The emergence of this model confirms that AI startups broadly face common challenges including scarce compute capacity, unstable GPU supply, financing difficulties and heavy cash flow pressure. This model offers a new pathway for resolving industry-wide liquidity pressure, making it worthy of in-depth research.

Disclaimer: The "Quick Summary" content is entirely generated by AI. Please exercise discretion when interpreting the information. For issues or corrections, please email run@ebrun.com .

I am a Brand Seller Factory Service Provider Marketplace Seller Researcher Read it again.

2026年7月2日,英伟达公布全新合作项目,面向高速增长的AI初创企业提供算力额度,换取对方未来产品及云业务的一定比例营收分成。英伟达将作为中间方,为参与项目的初创企业对接由英伟达芯片支撑的全栈计算资源。

参与该项目的云服务类AI企业、模型研发方及其他相关企业,均需按照约定与英伟达共享营收。英伟达披露的信息显示,首批为项目提供算力支撑的合作方共两家。澳大利亚算力服务商Sharon AI将部署最多4万台英伟达GPU。新加坡AI基础设施企业Firmus Technologies正在印尼巴淡岛建设数据中心,项目预计规模达360兆瓦,可容纳最多17万台英伟达GPU。

针对参与项目的云服务商,英伟达配套提供财务担保,若服务商无法获取足够AI开发者客户,英伟达将回租未被使用的GPU,承担对应未用算力的成本。该模式可同时覆盖GPU及数据中心的融资需求。这一动作也可降低英伟达对亚马逊、微软、谷歌等科技巨头的业务依赖,目前上述企业仍是英伟达芯片的主要采购方,同时也在推进自研AI硬件的相关布局。

当前AI行业内,算力稀缺是初创企业普遍面临的发展限制,GPU供给不稳定、成本波动较大,不少AI企业选择与芯片厂商达成营收或股权分成协议,缓解行业流动性压力。此前OpenAI已与亚马逊、AMD等多家合作伙伴达成多项协议,涉及股权收购及投资等合作内容。

本月早些时候,英伟达披露债务融资计划,知情人士透露募资规模至少200亿美元,募集资金将用于一般企业用途,包括现有债务的偿还及再融资。

文章来源:亿邦动力

广告
微信
朋友圈

FAQ回顾

英伟达算力换营收项目是什么?

2026年7月2日英伟达公布该合作项目,面向高速增长的AI初创企业提供算力额度,换取对方未来产品及云业务的一定比例营收分成,还为参与企业对接英伟达芯片支撑的全栈计算资源,为合作云服务商提供财务担保。

AI初创企业发展普遍面临哪些痛点?

当前AI行业内算力稀缺是初创企业普遍面临的发展限制,存在GPU供给不稳定、成本波动较大的问题,不少AI企业选择与芯片厂商达成营收或股权分成协议,缓解行业流动性压力。

英伟达推出算力换营收模式有什么作用?

该模式可同时覆盖GPU及数据中心的融资需求,还能降低英伟达对亚马逊、微软、谷歌等科技巨头的业务依赖,同时帮助AI初创企业破解算力不足的发展瓶颈。

英伟达算力换营收项目首批算力合作方有哪些?

首批算力合作方共两家,澳大利亚算力服务商Sharon AI将部署最多4万台英伟达GPU;新加坡AI基础设施企业Firmus Technologies正在印尼建设规模达360兆瓦、可容纳最多17万台英伟达GPU的数据中心。

这么好看,分享一下?

朋友圈 分享

APP内打开

+1
+1
微信好友 朋友圈 新浪微博 QQ空间
关闭
收藏成功
发送
/140 0