Nvidia Stock: BNP Paribas Raises Target to $345

Nvidia has received a higher price target from BNP Paribas, which raised its target to $345 from $285 and retained its "Outperform" rating. The revised target implies a potential upside of about 47% from the stock's Friday closing price, based on the brokerage's assessment.

The revised outlook comes as investors weigh Nvidia's position in the artificial intelligence (AI) chip market against growing competition from Advanced Micro Devices (AMD) and custom chips developed by major technology companies. BNP Paribas expects Nvidia to retain a strong position as AI workloads shift from training models to running them, a process known as inference.

The brokerage's assessment centres on Nvidia's combination of chips, networking hardware and software. It also expects the company's gross margins to remain above 70% despite rising competition.

Nvidia stock target price and valuation

BNP Paribas raised its Nvidia price target by $60, from $285 to $345, while maintaining its "Outperform" rating. Its valuation model uses 15 times calendar year 2028 earnings, plus cash, to arrive at the revised target. The brokerage's estimate reflects its expectations for Nvidia's earnings and competitive position over the coming years, rather than a guaranteed future share price.

The reported 47% upside is a projection based on the closing price used in the assessment. It is not a measure of the stock's actual return since the target revision. The available source also contains conflicting share price references: it says Nvidia had not crossed $300, while separately describing $237 as its previous all time high during the preceding week. These figures cannot be reconciled from the supplied information, so the precise reference price should be checked before using the upside estimate as a current market figure.

Particulars	Details
Brokerage	BNP Paribas
Rating	Outperform
Previous target price	$285
Revised target price	$345
Implied upside	About 47%
Valuation basis	15 times CY28 earnings, plus cash
Expected gross margin	Above 70%

Why BNP Paribas expects Nvidia to retain its advantage

Nvidia's position rests on more than the processing power of its graphics processing units (GPUs). Its CUDA software ecosystem and NVLink hardware architecture connect computing, software and data transfer within a broader system. BNP Paribas argues that this combination could be difficult for competitors to replicate with individual chips alone.

The distinction matters as AI systems become more complex. Training involves teaching an AI model using data, while inference happens when the trained model responds to requests or carries out tasks. Newer AI applications may need to reason through several steps, call external tools and work with longer amounts of context. Such tasks place demands on memory, networking and the movement of information across connected computing systems.

Nvidia is also developing an integrated approach to these requirements. The brokerage expects the company's full system architecture to remain an advantage as customers build larger AI infrastructure installations. This view is relevant to Nvidia's position in the AI infrastructure investment market, where spending extends beyond individual processors to the equipment needed to run AI applications at scale.

Vera Rubin platform puts focus on AI inference

Nvidia's upcoming Vera Rubin platform is another part of the brokerage's assessment. The platform combines GPUs, central processing units (CPUs), networking and other components in an integrated rack scale system designed for agentic AI.

Agentic AI systems are built to carry out tasks that involve several steps, make decisions within a workflow and use external tools. These systems can require repeated interactions between processors, memory and networking components. As a result, performance depends on how well the complete system works together, not simply on the speed of an individual chip.

BNP Paribas expects this architecture to help Nvidia defend its position as AI computing requirements change. The brokerage also sees networking as an increasingly important part of the business. When multiple GPUs operate within large computing racks, high speed connections help move data between components and keep workloads running efficiently.

The company's DLSS 5 launch for RTX graphics cards is a separate example of Nvidia's use of AI technology in its graphics business. However, the price target assessment described here is focused primarily on its AI computing systems, software ecosystem and networking capabilities.

Nvidia's expected share of the AI computing market

BNP Paribas expects Nvidia to retain at least 75% of the AI computing market by dollar value, even as AMD and custom accelerators gain ground. This is a brokerage forecast, not a confirmed future market share.

Custom accelerators are chips designed for particular computing workloads. Major technology companies have been developing their own processors to meet internal AI requirements. Such efforts could create competition for Nvidia, particularly when large customers seek more control over hardware costs and computing capacity.

However, BNP Paribas argues that Nvidia's integrated systems could help it defend its position. Its software ecosystem, interconnect technology and networking products give customers a connected computing platform. Rival chipmakers would need to compete across several parts of that system to challenge Nvidia's broader offering.

The brokerage's expectation of gross margins above 70% also reflects its confidence in Nvidia's pricing power and product mix. Actual margins will depend on future revenue, production costs, product demand and competition. The supplied material does not provide a detailed margin forecast by quarter.

Nvidia's latest financial results

Nvidia's reported second quarter results for the July fiscal quarter showed strong growth in AI related demand. Revenue more than doubled to $96.2 billion, while adjusted earnings reached $2.22 per share. The data centre business accounted for most of the revenue, highlighting the role of large scale AI computing in the company's financial performance.

Financial metric	Reported figure
Total quarterly revenue	$96.2 billion
Revenue growth	More than doubled year on year
Adjusted earnings per share	$2.22
Data centre revenue	$89 billion
Data centre revenue growth	117% year on year
Edge computing revenue	$7.2 billion
Edge computing revenue growth	27% year on year

Data centre revenue of $89 billion accounted for the largest share of Nvidia's reported quarterly revenue. The 117% year on year increase points to strong demand for the computing infrastructure used to develop and run AI systems. Edge computing revenue also grew, reaching $7.2 billion, although the supplied figures show that the data centre business remained the main revenue contributor.

Future results will help show whether this level of demand can be sustained as customers expand their AI infrastructure. Investors will also monitor whether revenue growth translates into margins above the level projected by BNP Paribas.

Nvidia expands share buyback programme by $150 billion

Nvidia recently increased its share repurchase authorisation by $150 billion. The company expects to complete the total remaining buyback programme through fiscal year 2028.

A share buyback allows a company to use its cash to repurchase its own shares. The number of shares outstanding can fall as repurchases are completed, although the actual effect depends on the number of shares bought and other changes in the share count. An authorisation sets the amount a company may repurchase, but it does not mean the full amount has already been spent.

The reported increase exceeded Apple's $110 billion expansion of its buyback programme in 2024, making Nvidia's additional authorisation the largest increase in a stock repurchase programme described in the supplied material. The buyback is separate from the brokerage's valuation model, which uses projected earnings and cash to calculate the $345 target.

Nvidia has also introduced an Open Agent Safety Platform for AI agents. The company said the platform combines its OpenShell and Sentry technologies to provide monitoring, policy enforcement and security controls outside the AI model itself. The announcement points to a separate area of development as AI systems become capable of carrying out more complex tasks.

Market analysis: competition and execution remain important

The outlook for Nvidia depends on whether demand for AI computing continues to grow and whether the company can defend its position as hardware requirements evolve. The shift towards inference could increase the importance of networking, memory and system level integration. Nvidia's broad platform is central to BNP Paribas' argument that the company can retain a large share of the market.

Competition remains a factor to watch. AMD and major technology companies developing custom accelerators could put pressure on Nvidia's market share or pricing. Customers may also change their spending plans as they assess the cost and performance of AI infrastructure.

The $345 target therefore rests on several expectations: Nvidia retaining at least three fourths of the AI computing market by dollar value, maintaining gross margins above 70% and delivering earnings that support the brokerage's valuation multiple. These are projections and may change as financial results and competitive conditions develop.

What investors should watch

The next set of Nvidia financial results will help test the assumptions behind BNP Paribas' valuation. Investors can monitor:

• Data centre revenue growth and demand from AI infrastructure customers.
• Gross margins and whether they remain above the brokerage's 70% expectation.
• Nvidia's market position as AMD and custom accelerators gain ground.
• Progress on the Vera Rubin platform and demand for integrated networking systems.
• The pace of share repurchases under the programme scheduled to run through fiscal year 2028.

The relationship between revenue growth, margins and competition will help determine whether the brokerage's earnings assumptions remain realistic. The $345 target is an analyst estimate, and the actual share price may move differently.

Frequently asked questions

What is BNP Paribas' new Nvidia stock target price?

BNP Paribas has raised its Nvidia stock target price to $345 from $285 and retained its "Outperform" rating. The revised target implies about 47% upside from the Friday closing price cited in the assessment.

How much upside does BNP Paribas see in Nvidia stock?

BNP Paribas sees about 47% potential upside in Nvidia stock based on its revised $345 target price. This is an estimate tied to the reference closing price, not a guaranteed return.

Why is BNP Paribas positive on Nvidia?

BNP Paribas expects Nvidia to benefit from its combined GPU technology, CUDA software ecosystem, NVLink architecture and networking products. The brokerage believes these systems could help Nvidia retain at least 75% of the AI computing market by dollar value.

What is Nvidia's Vera Rubin platform?

Nvidia's Vera Rubin platform is an integrated computing system combining GPUs, CPUs, networking and other components for agentic AI workloads. The platform is designed for systems that need to handle multistep tasks, use external tools and process longer contexts.

What were Nvidia's latest quarterly revenue figures?

Nvidia reported $96.2 billion in revenue for the July fiscal quarter, with adjusted earnings of $2.22 per share. Data centre revenue reached $89 billion, up 117% year on year, while edge computing revenue rose 27% to $7.2 billion.

How much has Nvidia added to its share buyback programme?

Nvidia increased its share buyback authorisation by $150 billion and expects to execute the total remaining programme through fiscal year 2028. The authorisation allows repurchases up to the approved amount but does not mean the entire sum has already been spent.