Nvidia Price Targets Are Spreading Out. Wall Street Is Pricing Very Different AI Outcomes

Nvidia's latest analyst actions show an unusually wide gap between firms that agree on the rating but disagree sharply on valuation. On 4 September, Rosenblatt reiterated a Buy rating with a $390 price target and Needham reiterated Buy with a $300 target, according to Benzinga's analyst-action database. The difference is notable because both firms remain positive on the stock while assigning very different values to that view.

The range becomes wider when the actions following Nvidia's 26 August earnings are included. Benzinga records Raymond James raising its target to $515, Evercore ISI to $465, Bernstein to $400 and a number of firms clustered around $300 to $350. These are dated analyst opinions, not guarantees or a single market consensus, and they should not be treated as interchangeable forecasts.

The earnings numbers explain the optimism, not the valuation gap

Nvidia reported fiscal second-quarter 2027 revenue of $96.2 billion, an increase of 106% from a year earlier. Data Center revenue reached $89.0 billion, up 117%. The company's SEC filing shows $48.7 billion of quarterly hyperscale revenue and $40.3 billion from AI clouds, industrial and enterprise customers within Data Center.

Those numbers make a bullish operating case easy to understand. They do not settle what multiple investors should pay for that growth. The difference between a $300 target and a target above $500 can reflect assumptions about AI infrastructure demand, gross margins, competition, customer concentration, future capital intensity and how rapidly today's growth rate normalises.

Why the range matters more than the average

A consensus target compresses different models into one number. That is convenient, but it can hide the part investors actually need to understand: where analysts disagree. A wide range after the same earnings report suggests the debate is no longer mainly about whether AI demand is strong. It is about the durability and valuation of that demand.

For a company already operating at Nvidia's scale, relatively small changes in long-term margin or growth assumptions can produce very large differences in estimated equity value. The upper end of the target range effectively assumes that Nvidia can convert a much larger AI infrastructure market into sustained earnings power without a proportionate erosion in economics. Lower targets can still be bullish while applying more conservative assumptions to that transition.

Our view: Nvidia is becoming a duration argument

The most interesting part of the current Nvidia debate is not that analysts disagree by hundreds of dollars. It is what that disagreement says about the stock. Nvidia increasingly trades as a judgment on the duration of the AI infrastructure cycle, not simply the next quarter's GPU shipments.

That makes the distribution of analyst targets more informative than a headline consensus figure. If revenue keeps compounding at extraordinary rates, the high-end models become easier to defend. If growth normalises faster, investors will discover that an excellent company and an excellent entry price are not the same thing. GMR will therefore track the sequence of target changes rather than treating one brokerage action as a definitive verdict.