Wedbush lifts Nvidia price target to $345 after first full-year guidance

Wedbush lifted its Nvidia price target to $345 from $330 and reiterated an Outperform rating after the company beat forecasts and issued its first full-year outlook. Q2 revenue rose 106% y/y to $96.2B (vs. $92.2B consensus) and adjusted EPS was $2.22 (vs. $2.10 expected). The forecast/beat combination supports upside sentiment and is likely to move NVDA shares modestly.

Analysis

The key market implication is not the upgrade itself; it is that the largest AI capex beneficiary is still extending visibility faster than sell-side models. That keeps the entire hyperscaler spending complex intact for the next 1-2 quarters and lowers the odds of a near-term digestion phase in the AI supply chain. The second-order winners are the constrained upstream enablers — advanced packaging, HBM, and foundry capacity — because sustained demand from NVDA makes allocation discipline more valuable than unit growth.

The losers are the “good enough” AI substitutes: AMD and, further out, custom silicon programs that need flawless execution to dislodge NVDA’s ecosystem advantage. If the market reads this as confirmation that NVDA is still taking share rather than merely raising prices, valuation dispersion should widen inside semis: NVDA multiple premium can hold while lower-conviction AI names face compression. By contrast, system integrators with thin margins can still lag if they need to carry inventory and pass-through costs before they monetize the buildout.

Risk is that this is a late-cycle confirmation event, not a new inflection. The next 1-3 month catalyst is whether customers translate the guide into incremental capex, and whether gross margin holds as product mix shifts; if margin or backlog commentary softens, the market will quickly treat the upgrade as stale. Over 6-18 months, the main falsifier is any evidence that AI monetization lags AI spend, because that would cap the duration of the semiconductor super-cycle and rotate leadership back to cash-flow names outside the AI chain.

Key Decisions for Investors

• Stay long NVDA on pullbacks rather than chasing strength; best entry is a 2-3% post-event fade, with upside to the next quarterly guide if hyperscaler capex remains intact. Risk/reward is favorable only if gross margin and forward commentary stay firm.
• Pair long NVDA / short AMD for a 1-3 month relative-value trade. Thesis: NVDA’s ecosystem and pricing power keep taking share, while AMD remains the most obvious beneficiary of any ‘good enough’ substitution narrative. Falsifier: AMD closes the product gap faster than expected or wins a meaningful design cycle.
• Add exposure to upstream AI bottlenecks via TSM and, for higher beta, MU as a proxy for HBM tightness. The trade works if NVDA’s demand signal keeps packaging and memory constrained for another 2-4 quarters; it breaks if customers signal inventory normalization or capex pause.
• Avoid chasing pure-play AI hardware laggards that depend on share gains, especially in semicap and server-assembly names, until there is evidence of broadening demand beyond NVDA. If the next print shows margin compression or weaker backlog conversion, those names are likely to underperform first.
• For options, prefer a modest NVDA call spread over outright calls for the next earnings window. The stock already embeds high expectations, so the best setup is limited downside if the post-upgrade move stalls, with upside only if management raises the forward path again.