In the crowded landscape of mesh networking hardware, big claims arrive with regularity. Most fade without much fanfare. But when a development team behind MeshTNC announced BYOMesh in early May 2026—a compact board packing both sub-GHz and 2.4 GHz LoRa radios—the promise of "up to 100×" more bandwidth than standard setups landed with unusual force.
The pitch is elegant, almost deceptively simple. Keep the sub-GHz radio doing what it does best: covering long distances across valleys and over ridgelines. Offload the heavy data lifting to a 2.4 GHz companion radio, sidestepping the power budget and complexity that comes with Wi-Fi or newer alternatives like HaLow. According to the announcement on partyon.xyz, this dual-radio architecture could transform how off-grid IoT networks handle backhaul.
Whether that 100× figure holds up in practice? That's where the skepticism begins.
Two Radios, One Board
At its core, BYOMesh pairs workhorses from Semtech's catalog: the SX1276 transceiver, a fixture in sub-1 GHz LoRa devices for years now, and the SX1281, which operates in the 2.4 GHz ISM band. The board itself is billed as a "companion dev kit"—small enough to slot alongside existing hardware and designed to run several popular mesh stacks out of the box. MeshCore, Meshtastic, MeshTNC, and Reticulum all get name-checked in the announcement, which was posted by nullagent and quickly circulated through aggregators like daily.dev and Hacker News.
The strategic thinking here isn't hard to grasp. By splitting access and backhaul across different frequency bands, developers gain flexibility—reach where you need it, throughput where it matters.
What's less clear are the finer details. Pricing? Still unconfirmed through public channels. PCB dimensions, connector types, enclosure compatibility, regulatory certifications—the dataparty shop lists the hardware, but specifications remain sparse. Perhaps the team is holding back until production stabilizes. Or perhaps they're letting early adopters figure out the details themselves.
The Bandwidth Claim, Unpacked
The 100× number is rooted in straightforward arithmetic—at least on paper. Semtech's datasheets show the SX1276 delivering LoRa bit rates around 37.5 kb/s, depending on spreading factor and bandwidth configuration. The SX1280/81 family, meanwhile, can theoretically push up to roughly 203 kb/s at 2.4 GHz under specific datasheet conditions: SF5, 1.625 MHz bandwidth, optimal RF environment.
That's a meaningful gap. When compared to ultra-robust sub-GHz configurations—think spreading factors cranked high for maximum range, data rates dropping to a few hundred bits per second—the multiplier starts to look plausible.
Then reality intrudes.
Regulatory constraints, for one. A GitHub issue opened in October 2025 within the MeshCore repository points to 47 CFR 15.247, the FCC rule governing the 902–928 MHz band in the United States. Under those regulations, certain sub-GHz LoRa configurations must meet a minimum 6 dB bandwidth requirement of 500 kHz unless they're using frequency hopping. That limits how conservatively you can tune the sub-GHz side without running afoul of compliance—and if your baseline isn't as slow as you'd like, the 100× multiplier shrinks fast.
Independent benchmarks and field tests remain absent. No lab reports have surfaced to validate the claim beyond what the datasheets suggest is theoretically achievable. The 100× figure appears to rest on modeling and use-case assumptions rather than measured performance. It's not implausible—a network using cautious sub-GHz settings for extreme range, layering in 2.4 GHz for relay-to-relay backhaul, might well see dramatic throughput gains—but it's not proven, either.
The community, predictably, is waiting for someone to actually run the numbers.
Where Would This Actually Work?

BYOMesh's backers point to Pacific Northwest long-range backhaul as a natural fit. Picture nodes perched on mountaintops, relaying sensor data or messages across forested valleys. Sub-GHz LoRa provides the reach—kilometers, sometimes tens of kilometers with line-of-sight and decent antenna placement—while 2.4 GHz handles the shorter but data-intensive hops between relay points.
Climate monitoring stations aggregating readings at a gateway. Disaster communication networks where a handful of strategically placed nodes need to shuttle traffic efficiently without relying on cellular or satellite backhaul. These aren't hypothetical scenarios. They're active areas of research and deployment.
An April 2026 preprint on arXiv explored dual-radio BLE and LoRa hierarchical meshes to offload traffic, a sign that mixing PHYs for capacity is attracting academic interest. A March 2026 preprint on low-cost LoRa mesh networks for large-scale environmental sensing underscored the continuing push to make these systems more capable without adding expensive infrastructure. Off-grid messaging, environmental monitoring, disaster resilience—these remain the core motivations driving LoRa mesh development through 2025 and into 2026.
The question isn't whether the use cases exist. It's whether BYOMesh delivers the performance those use cases demand.
An Ecosystem Play
One thing BYOMesh gets right: it's not trying to reinvent the software stack. By supporting MeshCore, Meshtastic, MeshTNC, and Reticulum, the board positions itself as an upgrade path for developers already embedded in these communities rather than yet another proprietary ecosystem demanding migration effort.
MeshTNC itself, developed by the dataparty organization, is an open-source tool for bridging LoRa mesh data with consumer radios and other transports. The broader dataparty footprint—MeshMarauder and related projects—suggests sustained commitment to LoRa mesh tooling, not a one-off experiment.
Meshtastic, perhaps the most visible player in this space, has seen rapid adoption. Mesh Radar, an open-source intelligence platform tracking nodes, claims to have indexed over 66,000 as of 2026. MeshCore, a newer entrant, is also gaining traction, judging by active development visible in its GitHub repository. By meeting these communities where they are, BYOMesh avoids the adoption friction that's killed plenty of promising hardware before it.
Skepticism in the Comments

On Hacker News and similar forums, reaction split predictably. Excitement from some—finally, a dual-radio board that doesn't require cobbling together separate modules. Skepticism from others, particularly around that 100× claim.
Several discussion threads highlighted the need for firmer substantiation, pointing readers to the MeshCore GitHub thread on US regulatory compliance. Achieving high throughput on the sub-GHz side isn't just a matter of turning knobs—there are legal and technical constraints, especially in the crowded 900 MHz band. Others questioned whether 2.4 GHz, already congested in urban environments, would deliver reliable backhaul where RF noise is a constant challenge.
The discussion reflects a community that's learned to be wary. LoRa mesh has long been a domain where tinkerers and serious infrastructure builders coexist, sometimes uneasily. Projects like "The Hammer," a 1-watt 900 MHz Meshtastic board from Broken Circuit Ranch, push the boundaries of what's possible with DIY long-range comms. BYOMesh enters a space where bold claims are expected to arrive with receipts attached.
Waiting for the Field Tests
For now, BYOMesh remains a promising datasheet and an intriguing architecture. The hardware is available for those willing to experiment—early adopters who don't mind working without comprehensive documentation or proven benchmarks.
Whether it delivers on its bandwidth ambitions will depend on factors the announcement barely touches: real-world RF conditions, regulatory compliance across different jurisdictions, interference mitigation at 2.4 GHz, power consumption under load, thermal management in compact enclosures. The list goes on.
The next chapter belongs to the engineers who'll put BYOMesh through its paces in forests, on rooftops, across disaster zones. Until those field tests arrive, the 100× claim remains exactly what it is: a theoretical ceiling, not a measured floor. Promising, maybe. Proven? Not yet.
