VendorsQCOMSEP 09, 2026

QCOM +9.5%: Amazon Warrant Ties Qualcomm to Up to $60B in AWS Chip Purchases Through 2036

Qualcomm issued Amazon a warrant for 25M shares at $161.26 vesting against up to $60B in server-chip purchases; revenue starts the December 2026 quarter.

BY GRETA REINHART · ENTERPRISE SAAS

Qualcomm disclosed in an 8-K filed with the SEC that it has issued Amazon.com NV Investment Holdings LLC a warrant for up to 25,000,000 shares of QCOM common stock at a $161.26 strike, vesting against up to $60 billion in AWS purchases of Qualcomm server chips, technology, systems, and manufacturing services through September 3, 2036. The face value of the warrant is roughly $4 billion; 3,750,000 shares vested at issuance against initial purchase commitments. QCOM rose as much as 9.5% intraday to $183.49, per TechTimes, its first positive year-to-date print, though CNBC noted a more modest close-of-tape move near 3%. Amazon slipped about 1%.

The structural read is what matters. This is Qualcomm’s first Western hyperscaler win, and it lands as a second inference supplier inside AWS alongside the incumbent Nvidia stack that has defined the buildout so far, including Nvidia’s $96.2B quarter and the capital cascading through neoclouds like Nscale’s Nvidia-anchored pre-IPO round.

The collaboration spans customized inference silicon across multiple chip generations plus 1.6 Tbps optical connectivity for intra-cluster networking. Qualcomm’s AI200 supports up to 768GB of LPDDR per card, a memory-capacity bet aimed at the decode phase of inference rather than HBM-bound training. That’s the technical tell: this is silicon designed for the per-query economics of models already in production, not for the next round of pretraining.

Speaking that morning at the Goldman Sachs Communacopia & Technology Conference, CFO Akash Palkhiwala said revenue begins in Qualcomm’s fiscal Q1 2027, the December 2026 quarter, with chips already in production. He held the $5 billion fiscal 2027 data center target with “very high confidence” and confirmed a second, unnamed global hyperscaler is engaged with similar scope.

Two caveats worth naming. The $60 billion is a warrant ceiling, not a signed revenue contract; remaining shares vest only against binding orders and actual purchases. And for anyone running AWS-hosted AI workloads inside 5–30 person teams, the operational implication is straightforward: a credible second inference supplier at the hyperscaler layer is the precondition for per-token prices to drift down. Build a re-pricing checkpoint into any 2027–2028 renewal.

Sources

About the author

Greta Reinhart

ENTERPRISE SAAS

Greta Reinhart tracks the enterprise software stack from San Francisco — data platforms, AI bundling, seat pricing, and channel checks across the largest SaaS vendors. She files on go-to-market shifts, packaging changes, and quarterly enterprise reads.