Week in Space Data, every Friday.
The dominant incumbent LEO broadband network versus Amazon's well-funded challenger: comparing constellation scale, technical approach, pricing, and commercial readiness.
Figures as of Aug 2026.
| Metric | Starlink | Project Kuiper |
|---|---|---|
| Operator | SpaceX (Starlink) | Amazon (Amazon Leo, formerly Project Kuiper) |
| Service Start | Public beta Oct 2020; full service 2021 | Prototype tests 2023; commercial 2025 |
| Satellites in Orbit (Aug 2026) | 10,000+ | Small in-orbit fleet: production launches began 2025 but the 50%-by-2026 FCC deadline was waived to 2029, so build-out is still early. See our live Constellations tracker for the current count. |
| Total Constellation Authorized | ~42,000 (FCC Gen 1 + Gen 2) | 3,236 (FCC licensed) |
| Orbital Altitude | ~340–570 km | ~590–630 km |
| Inter-Satellite Links | Yes (laser ISLs on V2 satellites) | Yes (planned on production satellites) |
| Launch Provider | SpaceX (Falcon 9 / Starship) | ULA Vulcan, Blue Origin New Glenn, Arianespace |
| Subscribers | ~12M (Q2 2026) | Not yet commercially available at scale |
| Typical Latency | 20–60 ms | Target <30 ms (per Amazon specs) |
| Download Speed (target) | 50–250 Mbps (consumer) | Up to 400 Mbps (per Amazon claims) |
| User Terminal | Starlink dish (phased array) | Amazon-designed phased array terminal |
| Consumer Pricing (US) | $120/mo (Residential) | Not yet announced |
| Parent Company Market Cap | SpaceX public (~$1.96T, SPCX, Aug 2026) | Amazon ~$2T+ (AMZN) |
| FCC Deployment Deadline | N/A (ongoing expansion) | Original 50%-by-2026 deadline granted a waiver extension to 2029 |
Figures as of Sep 25, 2026 (auto-updated daily from company profiles).
Starlink holds a commanding first-mover advantage with over 10,000 satellites in orbit and 4 million subscribers generating significant recurring revenue. SpaceX's vertical integration, building both rockets and satellites, gives it a structural cost advantage that no other operator can easily replicate. Starlink has demonstrated real-world performance across consumer, enterprise, maritime, aviation, and government segments.
Project Kuiper is backed by Amazon's balance sheet and benefits from integration with AWS cloud infrastructure, which may give it an advantage in enterprise and government cloud-connectivity use cases. Amazon has secured launch capacity across multiple providers (ULA Vulcan, Blue Origin New Glenn, Arianespace) to diversify risk. Kuiper's original FCC-mandated deadline required 50% of its 3,236-satellite constellation operational by mid-2026; that deadline has since been granted a waiver extension to 2029, easing near-term regulatory pressure but leaving the in-orbit fleet still in an early build-out phase: see our live Constellations tracker for the current satellite count.
Track both constellations on SpaceNexus
Yes. Amazon renamed Project Kuiper to Amazon Leo; the constellation, the FCC licence for 3,236 satellites and the launch contracts are unchanged. Older coverage still says Kuiper.
Starlink, by a wide margin: more than 10,000 satellites in orbit against an Amazon Leo constellation licensed for 3,236, of which only the early batches have launched. Starlink also has FCC authorisation for up to about 42,000.
Not at scale. Amazon Leo is in early deployment and initial service; consumer pricing has not been announced. Starlink residential service is available in most of the world at about $120 a month in the US.
Several providers: ULA's Atlas V and Vulcan, Blue Origin's New Glenn, Arianespace, and Falcon 9 (Amazon bought SpaceX launches when its other providers ran behind). New Glenn is grounded until its pad is rebuilt after the May 2026 explosion, which pushes more of the manifest onto the others.
The FCC originally required half the constellation in orbit by mid-2026; Amazon was granted a waiver extending that to 2029. Starlink has no comparable deadline because its network is already operating.