Russia is rolling out Rassvet, a low-Earth-orbit broadband constellation billed as a domestic answer to Starlink. In four months it has produced two serial launches, its first burned-up satellite, and assurances from the top that "the system exists and works." The reality is more modest: this is not a rival to Starlink, nor even a regional replacement for it, but the very start of a road on which the political showcase runs well ahead of the engineering.
Analytical review. Data as of 20 July 2026; some target figures are estimates subject to revision.
The big picture
Satellite connectivity has stopped being a civilian convenience and become a matter of military-strategic sovereignty — arguably the central technological lesson of the war in Ukraine. Elon Musk's Starlink showed that a low-orbit network can decide whether troops at the front stay in command of the battlefield, and that the off-switch sits with a private company in another country. It is this shift, rather than any abstract notion of "digital development," that explains why the Kremlin is pouring hundreds of billions of roubles into a constellation of its own.
Rassvet is best read as the extension into space of a logic Russia already applied on the ground: control over communications. The project is run by Bureau 1440, part of the same holding that built the country's "sovereign internet" and lawful-interception infrastructure. What was called control over traffic on the ground becomes, in orbit, control over the channel itself. In that sense Rassvet is less a commercial start-up than an instrument of the state's communications autonomy.
The trouble is that a wide gap separates the political layer from the engineering one — and that gap is the real story. As of July 2026 there are roughly two dozen working satellites in orbit; one from the first serial batch has already burned up in the atmosphere; and the lag behind Starlink is measured not in multiples but in hundreds. The claim that the system "functions" is formally true and, at the same time, misleading: what works is a demonstrator, not a service.
Two launches, and the silence around them
The timeline is short. In 2023 and 2024 Bureau 1440 put its experimental Rassvet-1 and Rassvet-2 missions into orbit — six satellites in all, used to trial the core technologies. It launched the first serial batch of 16 satellites on 23 March 2026 aboard a Soyuz-2.1b rocket. A second serial batch followed on 19 July 2026; the company did not disclose how many satellites it carried.
What stands out is less the cadence of the launches than the way they were handled. Eyewitnesses placed the March lift-off at the Plesetsk cosmodrome, yet neither Roscosmos nor the defence ministry made any official announcement — the first confirmation came in a video from the company itself, released only after the satellites had reached orbit. For a purely civilian telecoms venture, that secrecy is unusual. It can be read as an indirect sign that the programme is treated as strategic rather than merely commercial, even if officially Rassvet remains a broadband service.
By the company's own account, the serial satellites operate in the Ku and Ka bands, deliver speeds of up to 1 Gbit/s with latency below 70 ms, and carry in-house inter-satellite laser links, plasma thrusters and support for the 5G NTN direct-to-smartphone standard. These figures are the manufacturer's self-report; there is as yet no independent verification of throughput or latency under working conditions, and until the commercial service launches they are best treated as targets rather than proven performance.
Who is building Rassvet
Bureau 1440 grew out of the MegaFon 1440 project office, set up in November 2020. In 2022 it moved into the IKS Holding structure and changed its name; MegaFon stayed on as a strategic partner. The holding's pedigree matters more than the signage. IKS Holding was created by the entrepreneur Anton Cherepennikov — who died in July 2023 at the age of 40 — a long-standing business partner of Alisher Usmanov. The group takes in the computing-hardware maker Yadro, the R&D arm Kryptonite and, tellingly, the Citadel group.
Citadel, according to Meduza, Novaya Gazeta Europe and trade outlets, is Russia's largest maker of SORM lawful-interception systems and one of the chief beneficiaries of the "Yarovaya law" — the equipment through which the security services gain access to traffic and calls. The same holding that monetised ground-level control over communications is now building the orbital layer. This does not prove any military purpose for Rassvet, but it sets the frame: the project sprang not from civilian telecoms but from an environment for which communications are, above all, an instrument of state control.
The financial scaffolding is state-provided. The project sits within the "Data Economy" national programme: according to Kommersant and trade sources, some 102.8 billion roubles from the federal budget and about 329 billion roubles of the company's own funds are earmarked for the constellation through 2030. Both Prime Minister Mikhail Mishustin — who set a goal of conveyor-belt production of up to 250 satellites a year — and President Vladimir Putin, who called for a shift from experimental work to serial assembly, have publicly flagged it as a priority. The state here is not a distant sponsor but the customer.
The satellite that never climbed
The constellation suffered its first confirmed loss in early June. The satellite catalogued as NORAD ID 68363 — "Object 4" in independent trackers — from the March serial batch failed to reach its target altitude, lost contact and, around 6 June 2026, re-entered the dense layers of the atmosphere and burned up, less than three months after launch. The event was logged by the satellite-tracking sites n2yo.com and celestrak.org and by the independent observer Anatoly Zak; the company later confirmed the loss.
While the rest of the batch "rested," testing key systems before starting a planned orbit-raising, this craft could not hold its altitude and steadily sank. Bureau 1440 says six experimental Rassvet satellites and 15 of the 16 first-batch craft are operating normally in low Earth orbit. The trade publication ComNews framed it neutrally: nearly 90% of the first batch still stand a chance of reaching their working orbit.
The scale of the episode is easy to misjudge in either direction. On one hand, losing a single craft from a batch is statistically routine for low orbit: the far larger Starlink loses satellites regularly, and the very architecture of such networks builds in redundancy and spares. On the other, one failure out of 16 implies a very different reliability picture than one failure out of thousands, and at this early stage every satellite lost is an expensive lesson rather than background noise. The fair reading: not a disaster, but not the "trifle" the press office suggests either — it is the normal price of an early production run, and one more honestly left unmasked.
The Starlink lesson: why the Kremlin wants its own orbit
The project's strategic driver is not the market but dependency. Starlink proved its military worth on the Ukrainian front, delivering resilient connectivity in the combat zone — and, in the same breath, exposed the vulnerability of relying on it, since control of the network stays with the operator and its jurisdiction. Russian military channels complained of connectivity outages that they tied to terminals being switched off at the front — a subject FACTUM has examined separately — while the defence ministry insisted that Starlink restrictions had no effect on its forces. Both sides of that argument are parties to the conflict, and their claims are not independently confirmed; but the fact that the question became a matter of public dispute shows how sensitive it is.
Hence the political showcase. At a 12 June 2026 meeting with war veterans, Putin said Russia already had such a system and reduced the task to one of scaling up, describing it as "created" and functioning. That is formally true: demonstration satellites are in orbit and do carry traffic. But there is a fundamental difference between "the system functions" and "there is a deployed, military-grade service" — a difference the political phrasing erases. For now, any military use of Rassvet should be read as a declared aim and a reasonable inference from context, not as a confirmed operational capability.
Showcase versus Starlink: a question of scale
This is where the "Starlink analogue" rhetoric shatters on the numbers. By mid-2026 the active Starlink fleet had passed 10,000 craft, and SpaceX had launched more than 12,000 satellites in total since 2019 — over a thousand of them in the first months of 2026 alone. Rassvet has about two dozen working machines in orbit. The gap is measured not in multiples but in hundreds, and it is not narrowing: SpaceX launches batches of 20 to 30 satellites almost weekly.
Rassvet's own targets are best given as a range, since sources diverge. The federal programme, as reported, calls for the constellation to reach 156 satellites by the end of 2026; the commercial launch is slated for 2027, by which point the count is meant to exceed 250; full deployment — variously put at 288 to 292 satellites, spares included — is placed in 2027 to 2030, with a horizon of around 900 satellites for a global network by 2035. Even if these plans hold, Rassvet will remain an order of magnitude smaller than Starlink and will serve a different purpose: not global coverage but connectivity over Russian territory and, most likely, its zone of military interest.
From which follows the key point of calibration: Rassvet is better judged not as a "Russian Starlink" but as a bid for a minimally sufficient sovereign network. On satellite count the project loses by default; its point is not parity but possession of a channel that cannot be switched off from outside.
What the project genuinely gets right
For balance: behind the showcase there are real engineering achievements, and to ignore them would be as much an error as taking the political phrasing at face value. Bureau 1440 is a vertically integrated company that develops its own satellites, user terminals and ground infrastructure — something of an exception in the Russian space sector. By its own account it has proven inter-satellite laser links, transferring hundreds of gigabytes at around 10 Gbit/s between craft, along with in-house plasma thrusters and 5G NTN direct-to-smartphone connectivity, and it has opened automated production of solar panels.
The commercial contour is telling too: trials of satellite connectivity with Beeline have been announced, along with deployment on high-speed and long-distance trains. These are signs that the project is moving from lab to product, even if more slowly than advertised. In other words, the scepticism should attach to timelines, scale and the gap between rhetoric and reality — not to the plain fact that, for the first time in the post-Soviet era, Russia is building a serial low-orbit network, and building it on more than empty promises.
What could prove this assessment wrong
The cautious verdict — "a showcase for now, not a service" — would prove unfair under several scenarios.
First, if the production line genuinely delivers. The stated 10 to 12 launches a year, at roughly 15 satellites each, could within two or three years move the constellation from demonstration to working status; at that pace the hundred-fold gap ceases to be a meaningful argument, because a regional task does not need global scale.
Second, if the "minimally sufficient network" logic holds. Connectivity over Russia and the near-front zone does not require 10,000 craft; a few hundred at an 800 km orbit could give acceptable coverage, in which case comparing raw numbers with Starlink was always beside the point.
Third, patient state money. Unlike commercial ventures, Rassvet is under no obligation to pay back on schedule: if the political priority holds, the programme will survive technical setbacks that would have killed a private start-up.
Fourth, the statistical triviality of the loss. If the remaining craft reach orbit and see out their service lives, the burned-up satellite will stay a routine episode of an early production run rather than a symptom of systemic unreliability.
Finally, much of the key data is closed: the number of satellites in the July launch was not disclosed, and the military component is not officially confirmed. If a more mature programme lies behind that secrecy than is visible from outside, the public picture understates the real progress.
Conclusions: what is established and what remains hypothesis
Established. Rassvet is a real, state-funded low-orbit programme with two serial launches (March and July 2026), roughly two dozen working satellites in orbit and one confirmed loss — the first-batch craft that burned up on 6 June. It is run by Bureau 1440 of IKS Holding, a group historically tied to the infrastructure of state control over communications. Its lag behind Starlink on satellite count, as of mid-2026, runs into the hundreds.
Remains hypothesis. Any military purpose for Rassvet follows logically from the context and the secrecy, but is not officially declared or operationally confirmed. The stated technical specifications are the manufacturer's self-report, without independent verification. The constellation's target size and commercial-launch dates are planning figures that have already slipped once.
If current trends hold, 2027 and 2028 will settle the central question: whether Russia can turn a political declaration of sovereign connectivity into working infrastructure — or whether Rassvet remains an expensive showcase, behind which the strategic dependence on other people's networks does not go away. For now the cautious conclusion stands: in this field, sovereignty has been declared before it has been built.
Sources
Kommersant, Meduza, Habr, ComNews, Interfax, Forbes, RBC Trends, TAdviser, CNews, Novaya Gazeta Europe, dp.ru, ixbt, together with independent orbital-tracking data (n2yo.com, celestrak.org) and the observer Anatoly Zak; statements by Bureau 1440, Roscosmos and Russian officials. Estimates of satellite numbers and deployment timelines diverge between sources and are given as ranges. Data from one side of the conflict — statements by Russia's defence ministry and military channels on Starlink's effect at the front — is not independently confirmed. The constellation's technical specifications are given as stated by the manufacturer.