ALAB · Astera Labs, Inc. · Stock research

Current Thesis

The binary flagged in the July note resolved on 2026-08-04, and it resolved up. Q2 FY2026 revenue landed at $392.4M (+27% QoQ, +104% YoY) against ~$360.7M consensus, non-GAAP EPS $0.80 vs $0.69, and non-GAAP gross margin came in at 73.7% — above the ~73% guide that had been the open pricing-power question since the Q1 call. The guide is what re-set the story: Q3 revenue of $540–560M against $416.5M consensus, non-GAAP EPS $1.16–1.21 against $0.81, roughly 40% sequential growth at the midpoint, with Scorpio X-Series switches in volume production a quarter ahead of plan and management expecting it to be the largest product family by revenue in Q3. The narrative leg an investor is buying is the up-stack move from PCIe/CXL retimers into rack-scale switching, priced at higher dollar content per rack. What has not happened is price confirmation. Investing.com's coverage of the call recorded a 12.62% regular-session gain to $361.56 with shares easing 2.3% after hours to $353.34; the 2026-08-14 close of $321.61 sits roughly 11% under that, and 33.4% under the $483.02 52-week high. A P&L re-accelerating into a chart that has given back most of its post-print move is the setup as it stands.

Bullish and bearish views on Astera Labs, Inc.

The model's bull view on Astera Labs, Inc. (ALAB), in brief: Q2 FY2026 (2026-08-04 AMC): revenue $392.4M, +27% QoQ and +104% YoY, versus ~$360.7M consensus; non-GAAP EPS $0.80 vs $0.69 estimate; non-GAAP operating margin 39.1%, up 290bps sequentially. The bear view: Best guide in the company's history did not make a new high. Both cases follow in full.

Bull Case

• Q2 FY2026 (2026-08-04 AMC): revenue $392.4M, +27% QoQ and +104% YoY, versus ~$360.7M consensus; non-GAAP EPS $0.80 vs $0.69 estimate; non-GAAP operating margin 39.1%, up 290bps sequentially.
• The Q3 guide is the event, not the quarter. $540–560M against $416.5M consensus is a ~$130M gap to the Street at the midpoint; non-GAAP EPS $1.16–1.21 vs $0.81; GAAP EPS $0.87–0.92 vs $0.56. Guided non-GAAP operating margin ~43% on opex of $156–160M.
• Margin question answered for Q2: 73.7% non-GAAP gross margin cleared the ~73% guide, so the ~340bps step down from Q1's 76.4% read as mix rather than price concession.
• Scorpio X-Series entered volume production in Q2, one quarter early, with management citing over 10 customers engaged and several in pre-production or qualification on the 2026-08-04 call.
• Product-generation mix shifted fast: PCIe 6.0 products were over 50% of Q2 revenue versus roughly one-third in Q1.
• Targets re-rated in one session (2026-08-05): RBC $500 (Outperform), Evercore ISI $452, Jefferies $450 (from $270), Roth $450 (from $275), Citi $430 (from $315), Needham $425 (from $260), Morgan Stanley $382 (Overweight). Consensus average target of $391.77 across 26 analysts (S&P Global) now sits above the 2026-08-14 close of $321.61, having sat at $272.47 — below spot — in late July.

Bear Case

• Best guide in the company's history did not make a new high. Eight sessions after the print the stock is ~11% below the level Investing.com recorded around the report and a third below the June peak. Buyers took the news and sold the strength.
• Q3 gross margin is guided to ~72%, down from Q2's 73.7%, explicitly attributed to mix toward lower-margin switching and module revenue. The up-stack move that raises dollar content per rack dilutes margin by construction, and each further leg into switches and optics repeats the trade-off.
• Not every desk bought it: TD Cowen cut its target to $375 from $425 on 2026-08-05 and kept a Hold rating.
• Insider distribution happened at higher prices. Board chairman Manuel Alba sold ~$60.5M on 2026-07-01 under a 10b5-1 plan; No new Form 4 activity has appeared in the filings feed since the print, so whether distribution resumes once the post-earnings window reopens is unresolved.
• Beta near 3.67 makes this the most amplified expression of the AI-connectivity cohort. SOXX logged its worst month since 2002 per 2026-08-03 coverage quoting Ed Yardeni; a repeat of the 2026-07-16/17 Samsung/SK Hynix-led slump takes ALAB down multiples of the group.
• Valuation still depends on the guide arriving. The trailing multiple stood near 197x at the July lows; the Q3 EPS step-up compresses it only once it is reported in November. Between now and then the stock carries the multiple, not the earnings.

Setup & Price Structure

• Reference: the 2026-08-14 close of $321.61. RSI(14) 58.2 — mid-range, neither extended nor washed out. Three-month return +38.2%. Distance from the $483.02 52-week high: -33.4%.
• The post-print band is roughly $291 to $362. The floor is the 2026-07-24 close of $291.58 (-10.82% that session), which broke the June recovery shelf; the ceiling is the $361.56 level recorded around the 2026-08-04 report. Price sits near the middle, about 10% above the July breakdown shelf.
• The narrative is accelerating, dated to 2026-08-04 (Q3 guide ~30% above consensus, Scorpio X pulled in a quarter) and 2026-08-05 (seven-plus target raises in a single session, consensus average target moving from $272.47 to $391.77). The qualifier matters: the fundamental leg accelerated and the price has not yet followed. If the $291–$362 band resolves lower rather than higher, the label belongs in maturing.
• Crowding and positioning observables, stated as observables: consensus average target now sits ~22% above the last close, so the sell side is chasing rather than capping; the coverage cluster on 2026-08-05 was one-day and broad; RSI at 58.2 shows no momentum extreme; the post-print fade of ~11% over eight sessions indicates supply into strength; June–July insider sales at $60.5M scale occurred 12–25% above the current close; Nasdaq-100 inclusion effective 2026-06-22 mechanically broadened the passive holder base.
• No company-specific event sits inside the next 30 days. The nearest dated driver is external (see below), which means the band is likely to be resolved by cohort tape rather than by ALAB news.

Catalyst Calendar (next 30 days)

• 2026-08-26 — NVIDIA Q2 FY2027 (AMC), guided ~$91.0B ±2%. Rack scale-up connectivity demand is downstream of NVL rack shipments; a cadence or capex signal here moves ALAB in the same session, and it is the only dated hard catalyst inside the window.
• ~2026-11-03 (est.) — Q3 FY2026 print. Outside the 30-day window, but the first actual test of the $540–560M revenue guide and the ~72% gross-margin guide. Nothing between now and then verifies either number.

Elapsed catalysts

• ~2026-09 (est.) — Q3 2026 investor-conference participation announcement. The company published a Q1 schedule (Jan 14 New York, Mar 3 San Francisco) and a Q2 schedule (May 19 J.P. Morgan, May 27 TD Cowen, Jun 3 Evercore); no Q3 2026 dates were confirmed as of 2026-08-16. (passed 10d ago)

What Would Change Our Mind

The structural break is the guide arriving as mix dilution without the volume that justifies it: Q3 revenue landing at the $540M low end while gross margin prints under the ~72% guide, which would say the switching and module ramp buys revenue at a worsening price. A second break is external — an NVIDIA print on 2026-08-26 that flags rack-shipment slippage or trims the ~$91.0B ±2% framework would hit a beta-3.67 downstream supplier harder than it hits NVIDIA. A third would be renewed officer or director selling at the $60M scale seen on 2026-07-01 once the post-earnings window reopens, which would argue the June–July distribution reflected a view rather than a calendar. On price, a weekly close below $291 erases the post-Q2 re-rating and returns the chart beneath the 2026-07-24 breakdown shelf; at that point the 2026-08-04 guide would have failed to hold a bid for a full month and the accelerating label no longer describes what is happening.

Correlation Notes

• The July drawdown was cohort-wide — Samsung, SK Hynix and KOSPI weakness on 2026-07-16/17 bleeding into US semis — not a company-specific stumble; ALAB fell with the group and has recovered with it ("Chip Stocks Find Buyers After Earnings Shock Shakes Out AI Trade", 2026-08-07).
• Beta near 3.67 makes the name a leveraged read on AI-capex sentiment: direct correlation to NVDA, AVGO and MRVL, and to SOXX, which per 2026-08-03 coverage had just logged its worst month since 2002.
• Rate sensitivity is now visible in the tape: the 2026-08-07 weak-payroll session was listed as an AI and tech rally, so long-duration semis are trading against Fed expectations alongside capex news.
• Index-flow correlation to the Nasdaq-100 stepped up mechanically with inclusion effective 2026-06-22, which links the name to broad-index rebalancing as well as to sector rotation.