The editorial argues every design choice — dual-flash unbrickable boot, mikroBUS expansion header, JTAG breakout, screws instead of clips, no captive app or cloud account — reveals hardware made by people who actually read and write firmware. This inverts the twenty-year consumer router contract where vendors ship locked-down SDK firmware and OpenWrt reverse-engineers support afterward.
By submitting the official OpenWrt One hardware page to Hacker News, the submitter frames this as a milestone worth surfacing to the developer community — an open hardware router blessed by the firmware project itself rather than yet another consumer SKU needing reverse engineering.
The bootloader-on-NOR / OpenWrt-on-NAND split guarantees a recovery path so experimentation can't brick the device, while the exposed JTAG header and mikroBUS Click slot invite hardware modification — LTE modems, LoRa radios, sensors. These are features that only matter to people who intend to open the case and hack on it.
The editorial highlights that the 202-point front-page thread is 'unusually free of the usual cool but I'll wait for v2 chorus' and that people are actually ordering the device. This reaction suggests the market has been quietly waiting for someone to solve the OpenWrt-compatibility-lottery problem with first-party hardware.
OpenWrt — the Linux distribution that a generation of network nerds has been flashing onto consumer routers since 2004 — now ships its own hardware. The OpenWrt One, developed in collaboration with the Software Freedom Conservancy and manufactured by Banana Pi, is the project's first officially blessed router. It retails for around $89, uses a MediaTek MT7981B SoC (dual-core Cortex-A53 at 1.3 GHz), 1 GB of DDR4, 128 MB SPI-NAND plus 128 MB SPI-NOR, dual-band WiFi 6 (AX3000-class), one 2.5 GbE WAN port, one 1 GbE LAN port, USB 2.0, and — the part that tells you who this was really built for — a mikroBUS header, a real-time clock with battery backup, and a JTAG debug port broken out on the PCB.
This is a router designed by the people who write the firmware, for the people who read the firmware. Every design decision reads that way. The dual-flash layout means the bootloader lives on NOR while OpenWrt itself lives on NAND, so a corrupted image can't brick the device — you always have a recovery path. The mikroBUS slot lets you snap in a Click board (LTE modem, LoRa radio, environmental sensor, whatever) without touching the case. The case itself is a simple metal enclosure with screws, not clips. There is no captive vendor app. There is no cloud account. There is a physical reset button that does exactly what it says.
The hackernews thread (202 points, top of the front page at time of writing) is unusually free of the usual "cool but I'll wait for v2" chorus. People are ordering them.
For twenty years the consumer router market has run on the same busted contract: a vendor ships hardware with a locked-down, security-abandoned SDK-derived firmware, and OpenWrt reverse-engineers support in months or years later — sometimes never. Buyers picked routers by cross-referencing the OpenWrt Table of Hardware against Amazon listings, praying the SKU they clicked was the one with the compatible flash chip and not the silent revision-B respin with a different SoC. The OpenWrt One inverts that: the firmware team specified the silicon, so day-one support is a given, and "day one" lasts as long as the project does.
The economic story here is more interesting than the technical one. A $89 router with these specs is not remarkable — you can buy a GL.iNet Flint 2 or a Xiaomi AX3000 for similar money with roughly similar silicon. What's remarkable is that a nonprofit software project raised the capital, negotiated the manufacturing run, and shipped a physical product without a VC round or a Kickstarter cliffhanger. The Software Freedom Conservancy took a small per-unit royalty; the rest is a genuinely community-built device. If this sells through, it's a template other open-source projects can copy. Pi-hole hardware. Home Assistant hardware. Nextcloud hardware. The pattern is now proven.
The mikroBUS choice deserves a second look. mikroBUS is a MikroElektronika-defined 16-pin footprint that exposes SPI, I2C, UART, PWM, analog in, interrupt, reset, and power — roughly, everything a small peripheral needs. There are hundreds of Click boards in the wild. Slap a cellular modem Click into that slot and you have a $130 failover WAN. Slap a LoRa Click in and you have a router that also runs a Meshtastic node. None of this requires soldering, opening the enclosure, or writing a custom driver — the mainline kernel already knows how to talk to most of these parts.
The community reaction is worth reading. The HN thread is full of engineers pointing out things vendors quietly gave up on: the RTC means your firewall rules with time-of-day logic work after a power outage without waiting for NTP. The JTAG header means you can debug a kernel panic instead of RMAing the device. The 2.5 GbE WAN is deliberately paired with 1 GbE LAN because the target user is someone on a gig-plus fiber plan whose LAN traffic is dominated by wireless clients — an honest read of what home networks actually look like in 2026, not a spec-sheet flex.
If you run a homelab, this is probably the router you should be running. Not because it's the fastest — it isn't — but because it's the one that will still get security patches in 2031, and the one where you can actually audit what's between your LAN and the internet. The threat model of a consumer router in 2026 is not "someone brute-forces WPA2"; it's "the vendor ships a compromised update, or stops shipping updates entirely and leaves a known CVE unpatched for years." The OpenWrt One removes both of those failure modes by construction.
For small-business and edge-compute use cases, the mikroBUS slot changes the calculus. A branch office deployment that used to require a $400 router plus a $200 LTE dongle plus a USB hub plus cable management now fits on one board. For network engineers who've been building custom edge boxes on Protectli or PC Engines APU2, this is a cheaper alternative for the low end — you give up x86 flexibility and get purpose-built hardware with a hardware crypto accelerator that actually gets used by the mainline OpenWrt WireGuard and IPsec stacks.
The caveat: this is not a router for someone who wants to unbox it and forget about it. There's no mobile app. The web UI (LuCI) is functional and improving, but it's not "eero" polished. If your users include a spouse who is going to call you when the WiFi is slow, plan for that. The upside is that when it does go wrong, you have a real Linux system with real logs and a real shell, not a black box.
The bigger question the OpenWrt One raises isn't about routers — it's about whether "open-source project ships its own hardware" becomes a repeatable pattern or a one-off curiosity. The economics only work if a critical mass of the community actually buys the thing instead of just starring the announcement. Early signs are good; the first production run reportedly sold out. If Banana Pi commits to a second run and OpenWrt uses the royalties to fund a v2 with WiFi 7, the pattern is established. If it stalls, the lesson is that even the most sympathetic open-source audience will still buy the $60 TP-Link and flash it themselves. Either way, for the next twelve months, this is the most interesting $89 in home networking.
Off topic, but what amuses me about the "Wrt" name is that it was originally alternate firmware for the Linksys WRT54G router from 25 years ago. The name has stuck for whatever reason; I guess since only geeks use it and know what it is.
What a coincidence to see this on the front page!I just received my OpenWrt One because I’m tired of dealing with the questionable quality of most routers.And I don’t feel like resurrecting my old PC that I used as a router for a while. I stopped doing that because it’s loud. Pretty sure the power s
How about OPNSense on open hardware of your choice, and passing messy wireless to separate AP?OpenWRT is very good, but the installation and upgrades are not easy. There is a zoo of images for different hardware, installation options and tools. It has to run on small devices, so there are limitation
If someone is interested in open hardware running openwrt, also check out Turris.https://www.turris.com/en/I have been using them for years and I'm really happy. I recently bought the WiFi 6 upgrade kit for both of my turris. They "recently" released their latest v
Top 10 dev stories every morning at 8am UTC. AI-curated. Retro terminal HTML email.
They are working on an OpenWRT Two at the moment which will be Wifi 7.OpenWRT runs on a lot of hardware and its a great way to extend the life of a router past the manufacturers patches as well as gain a lot of capabilities. I wouldn't buy a commercial router that wasn't supported by OpenW