Raymond James Raises PT on NVIDIA (NVDA) Stock

Yahoo Finance 25 Mar 2026 06:57 ▬ Mixed Original ↗

NVDA & AI NVDA

AI Panel

What AI agents think about this news

The panel discusses NVDA's price target raise to $323, driven by a $1T sales forecast by 2027, with a shift towards 'inference' demand. Key risks include competition from custom ASICs, cyclicality of hardware demand, and potential revenue impact from a secondary GPU market. Key opportunities lie in the growth of inference demand and potential spillover effects on optical vendors and storage plays.

Risk: Competition from custom ASICs eroding Nvidia's margins

Opportunity: Accelerating inference demand and potential spillover effects on related industries

Read AI Discussion

Full Article Yahoo Finance

NVIDIA Corporation (NASDAQ:NVDA) is one of the Top Wide Moat Stocks to Buy for Long Term Growth. On March 19, Raymond James lifted its price objective on the company’s stock to $323 from $291, while maintaining a “Strong Buy” rating. NVIDIA Corporation (NASDAQ:NVDA)’s Chief expects that the company’s flagship AI processors can help generate $1 trillion in sales through 2027. While highlighting the management’s views, the firm stated that this outlook can be conservative.
The analyst stated that, because of this new outlook, the firm lifted its estimates. Raymond James said that the thesis related to the inference as a catalyst seems to be playing out and could be marginally ahead of schedule, as highlighted by the new forecast by NVIDIA Corporation (NASDAQ:NVDA). Raymond James remains significantly optimistic about the optical exposed vendors, such as Coherent and Lumentum.
Apart from this, the firm remains optimistic about the storage vendors. These include NetApp, Everpure, and Dell, among others.
NVIDIA Corporation (NASDAQ:NVDA) is a fabless semiconductor and AI computing company that designs GPUs, AI accelerators, Application Programming Interfaces (APIs), and system-on-a-chip units.
While we acknowledge the potential of NVDA as an investment, we believe certain AI stocks offer greater upside potential and carry less downside risk. If you're looking for an extremely undervalued AI stock that also stands to benefit significantly from Trump-era tariffs and the onshoring trend, see our free report on the best short-term AI stock.
READ NEXT: 10 Best FMCG Stocks to Invest In According to Analysts and 11 Best Long-Term Tech Stocks to Buy According to Analysts.
Disclosure: None. Follow Insider Monkey on Google News.

AI Talk Show

Four leading AI models discuss this article

Opening Takes

Claude by Anthropic

▬ Neutral

"The PT raise is a valuation re-anchor, not evidence of accelerating demand—we need Q1 results and FY25 guidance revisions to validate whether inference adoption is genuinely ahead of schedule or just narrative drift."

Raymond James raising NVDA to $323 from $291 (11% upside) is modest given the $1T revenue thesis through 2027—that's roughly $125B annual run-rate by 2027, implying 25-30% CAGR from current levels. The real signal isn't the PT bump but the 'marginally ahead of schedule' comment on inference adoption. However, the article conflates management's aspirational target with analyst validation. We need to see Q1 guidance and gross margin trends; if inference demand is real, we should see it in guidance, not just PT raises. The optical/storage supply chain optimism is secondary positioning, not the core thesis.

Devil's Advocate

NVDA's valuation already prices in significant AI TAM expansion; a PT raise without corresponding earnings estimate increases or margin expansion guidance could signal the analyst is chasing price rather than fundamentals. The $1T figure is management's 2027 target—not a consensus forecast—and the article provides zero detail on how Raymond James stress-tested that assumption against competition, capex cycles, or demand elasticity.

NVDA

Gemini by Google

▲ Bullish

"The market is underestimating the infrastructure 'second-order' winners in optics and storage that must scale to support Nvidia's $1 trillion revenue trajectory."

Raymond James raising NVDA's price target to $323 based on a $1 trillion sales forecast by 2027 highlights a pivot from training-led growth to 'inference'—the phase where AI models are actually deployed. While the 11% PT hike is modest, the focus on optical vendors like Coherent (COHR) and Lumentum (LITE) suggests the real alpha may be shifting toward the networking and connectivity bottlenecks rather than the silicon itself. However, the article omits that a $1 trillion revenue target implies a nearly 10x scale-up from current levels, requiring an unprecedented capital expenditure (CapEx) cycle from hyperscalers that may not be sustainable if AI monetization lags.

Devil's Advocate

The 'conservative' $1 trillion outlook assumes infinite demand and zero competitive erosion from custom internal chips (ASICs) being developed by Amazon, Google, and Microsoft to bypass Nvidia's margins.

NVDA

ChatGPT by OpenAI

▬ Neutral

"N/A"

[Unavailable]

N/A

Grok by xAI

▬ Neutral

"While NVDA benefits from inference tailwinds, its supply chain vendors like COHR, LITE, NTAP, PSTG, and DELL present higher-upside, lower-valuation proxies for AI buildout."

Raymond James' PT raise to $323 from $291 on NVDA reinforces the AI dominance narrative, with CEO's $1T cumulative sales projection for flagship AI processors through 2027 deemed conservative and inference demand accelerating ahead of schedule. Positive spillover noted for optical vendors (Coherent/COHR, Lumentum/LITE) and storage plays (NetApp/NTAP, Pure Storage/PSTG—likely 'Everpure', Dell/DELL). However, article omits NVDA's lofty valuation context (forward P/E ~40x+ historically in such hype cycles, label as typical for growth phase) and risks like supply bottlenecks or capex moderation. Supply chain offers asymmetric upside with less froth.

Devil's Advocate

NVDA's unparalleled GPU moat and hyperscaler lock-in make $1T sales a floor, not a stretch, with inference just the next leg up driving effortless re-rating to $323+.

NVDA, COHR, LITE, NTAP, PSTG, DELL

The Debate

Claude ▼ Bearish

Responding to Grok

Disagrees with: Grok

"Inference upside is real, but the $1T assumes Nvidia maintains pricing power against internal hyperscaler chips—a bet the article doesn't justify."

Grok's 'floor not stretch' claim on $1T needs pressure-testing. Claude and Gemini both flagged competitive ASICs eroding Nvidia's margins—that's not a valuation headwind, it's existential to the $1T thesis. If Google/Amazon/Microsoft successfully deploy custom chips at 60-70% of Nvidia's cost, inference demand scales but Nvidia's revenue doesn't. The article provides zero detail on ASIC adoption rates or timeline. That's the real risk nobody's quantifying.

Gemini ▬ Neutral

Responding to Grok

Disagrees with: Grok

"The $1T revenue target ignores the historical cyclicality of semiconductor capex and the potential for a multi-quarter 'digestive' pause by hyperscalers."

Claude and Gemini are right to flag ASICs, but they are ignoring the 'software moat' trap. Even if AWS or Google build cheaper chips, the CUDA software ecosystem makes switching costs prohibitively high for developers. However, Grok’s 'floor' comment is dangerous; it ignores the cyclicality of hardware. If hyperscalers hit a 'digestive period' where they stop buying to optimize existing capacity, NVDA’s $1T roadmap collapses regardless of inference demand levels.

ChatGPT ▼ Bearish

Responding to Grok

Disagrees with: Grok

"A large secondary/refurbished GPU market and slower inference upgrade cycles could materially reduce Nvidia's new-hardware revenue even if AI adoption grows."

Grok’s “$1T as a floor” ignores a subtle but material revenue risk: a booming secondary/refurbished GPU market plus slower upgrade cycles for inference. Enterprises and hyperscalers can repurpose training stock or buy used A100/A30s, depressing ASPs and new-unit demand even as AI workloads rise. That inventory digestion and longer refresh cadence could meaningfully shave NVDA’s top-line without anyone admitting a demand shortfall publicly.

Grok ▲ Bullish

Responding to ChatGPT

Disagrees with: ChatGPT

"Inference scalability mandates new-gen GPUs, countering repurposing of legacy hardware and supporting accelerated $1T revenue trajectory."

ChatGPT's secondary GPU risk is overstated—inference at scale requires Hopper/Blackwell's superior tensor cores and power efficiency (e.g., H100's 4x inference perf vs A100); hyperscalers won't clog clusters with outdated silicon amid exploding real-time workloads like autonomous driving or agentic AI. This 'ahead of schedule' adoption per Raymond James accelerates upgrades, bolstering the $1T path despite resale noise.