MeshCore Firmware Distributions
Accurate as of 13 July 2026. Firmware projects movemoves fast. EveryCheck versionversions number,and device count and current figure below should be re-checkedcounts against the configurator and each project's releases page before you relyrelying on it.them.
ThereTwo aredifferent twochoices separateget questionsconfused towith answereach before you flash a MeshCore node, and people often confuse them.other.
Which distribution do you flash?distribution? That is what this page covers.page. Stock MeshCoreMeshCore, isor one option.of Severalfive independentcommunity projectsbuilds also build MeshCore firmware,that each withdo something different: save power, add a differenttouchscreen, priority:make lowermessages powermore draw,likely ato touch UI, better message delivery, a different operating system underneath. Some are forks of the upstream code, some are ports, and two are separate applications that link MeshCore as a library. "Distribution" is our umbrella word for all of them.arrive.
Which role does the node run?role? Companion, Repeater, Room Server and so on.Server. That is covered on MeshCore Firmware Variants Explained. Pick a distribution first, then a role withininside it.
TheThese six distributions here are the ones offered by the Mesh America Device Configurator. There are other MeshCore forks in the wild; these are the ones you can flash from ourthe tooling.Mesh America Device Configurator.
First:You nonecan ofalways thischange isyour permanentmind
Flashing is reversible. If you do not like a distribution, flash a different one. You are not going to brick ayour node by changing firmware through the web flasher,radio, and goingyou can always go back to stock is always available.stock.
What you can lose is your statenode identity and contact list. A node's identity keypair, its contact list and its saved radio settings live in flash storage, and someclean install pathsusually wipe that storage while others preserve it. A full or "clean" install typically resetsgives the node to a new identity, which meansso your contacts will see you as a new nodeperson and will have to re-add you.you Assumeagain. a distribution change resets your identity unless the project says otherwise, and writeWrite down your radio settings (frequency, bandwidth, spreading factor, coding rate) before you start so you can put them back.start.
Start here before you flash anything:See Flashing MeshCore Firmware, andor Flashing OTA for a node you cannot physically reach, Flashing MeshCore Firmware OTA.reach.
Do they all interoperate?work together?
That is the question everyone asks, so here is a careful answer rather than a comfortable one.
All sixThey are builtall onmeant the MeshCore protocolto, and all six are intended to interoperate on the same mesh. Eacheach project says so about itself: ZephCore describes itself as "aiming for full protocol compatibility" with the Arduino firmware and the mobile apps, WADAMESH says it usesdoes. "theNobody same MeshCore protocol and phone-app companion as stock", and MCLite says it is compatible with other MeshCore devices and the official apps.
Those are the projects' own claims about themselves. We are not aware of any independent,has published interoperabilitya test acrossproving all six,it, and we have not run one.one, Treatso thetreat general claimit as well-foundedvery butlikely notrather proven.than certain.
There is at least one knownOne case wheredoes compatibilitybreak: genuinelyKeymind breaks,Cascade sends bigger "path hashes" by default, and itnodes is not hypothetical:
path.hash.modepath.hash.mode=2What else varies between distributions is settings, defaults, and how much airtime a node spends. Those affect the mesh even when every packet is valid.deploy.
The six at a glance
| Distribution | What it is | Pick it when |
|---|---|---|
| MeshCore Official MeshCore core team |
The Roles: Companion, Repeater, Room Server, GUI, KISS |
|
| EasySkyMesh PowerSaving IoTThinks |
Roles: Companion (BLE), Repeater, Room Server |
Solar or battery repeaters, where |
| Keymind Cascade mikecarper |
Roles: Companion (BLE, USB, Wi-Fi), Repeater, Room Server, Terminal Chat, Sensor |
Your |
| MCLite laserir |
Roles: Companion only |
You |
| WADAMESH ALLFATHER BV |
A Roles: Companion only |
You have a touchscreen |
| ZephCore liquidraver |
A Roles: Companion (BLE, USB, TCP), Repeater, Room Server, Observer |
Each distribution ships its own selection of roles. Some build roles the official flasher does not (Keymind Cascade builds Sensor, Terminal Chat and Companion Wi-Fi; ZephCore builds a TCP companion and an Observer). It is not a strict subset of the official set.
MeshCore Official
The upstreamstandard project,firmware, at github.com/meshcore-dev/MeshCore, created by Scott Powell (Ripple Radios) and now maintained byfrom the MeshCore core team. MIT licensed. This is what flasher.meshcore.io serves,gives andyou. whatEverything theelse Meshon Americathis configuratorpage serves under "MeshCore Official".
Start here. Companion (BLE), Companion (USB), Repeater, Room Server, GUI (and a GUI-with-SD variant) for screen-equipped boards, and a KISS radio build. Seven role keys, exactly.
NoteTwo onthings Sensor firmware. The upstream repository contains a simple_sensor example application, so the source exists. The official flasherit does not buildgive it:you. thereThere is no Sensor rolebuild and no Wi-Fi Companion build in the flasherofficial catalog,flasher. andThe nocode sensorfor binariesboth inexists theupstream, upstream releases. Ifbut you wanthave ato prebuilt sensor binary you either buildcompile it yourself or take one fromget a forkready-made (Keymind Cascade ships them). See MeshCore Sensor Nodes for how sensor telemetry actually works.
Note on Wi-Fi companion. Same story. Wi-Fi companion is a real, separate build, but the official flasher does not ship it, because the SSID and password are compiled in. You build it yourself or take a prebuilt binarycopy from a fork. Keymind Cascade prebuilds both.
EasySkyMesh PowerSaving
MaintainedA byversion IoTThinks at github.com/IoTThinks/EasySkyMesh, with the source changes in github.com/IoTThinks/MeshCore. The project's own description:
PowerSaving firmware isof MeshCorefirmwaretunedthattohasdrawbeenlesshighlypower,optimizedfromforIoTThinks.lowerSamepowerfeatures,usage.longerFunctionallybatteryitlife.works the same as MeshCore, and eventually many of the changes make their way into the MeshCore official firmware.
That last clause is the project's stated intent. We have not independently confirmed which of its changes have actually been merged upstream, so read it as an aim rather than a record.
About the power"15 numbers.mA" headline. TheIts PowerSaving15own releasetest table is titledmore "15mAvaried for ESP32 BLE Companions and no time-drift for repeaters". Readthan the release'sheadline own table before you budget a solar install on that headline: it reportssuggests: Heltec v3 at 19.6 mA, Heltec v4.3 at 24.9 mA with the front-end module on and 18 mA with it off,mA, Xiao S3 at 16.3 mA, and Xiao C3 at 15.1 mA. Only the Xiao C3 actually lands nearhits 15 mA. TheReal "nosavings, timebut drift"plan fixyour issolar taggedaround BETAthe innumber thatfor sameyour release.board, not the headline. These are the project's own measurements; we know of no independent verification. The direction is rightmeasurements and thenobody workelse ishas real,checked but size your battery off the figure for your board, not the headline.them.
PowerSaving16More: hasEasySkyMesh. alsoNot picked up the outpath return-path fix describedavailable in the Keymind Cascade section below, so some of the deliverability work is reaching this fork too.
See also EasySkyMesh: Third-Party Power-Optimized MeshCore Fork. EasySkyMesh is not available through the official MeshCore flasher.
Keymind Cascade
A fork of meshcore-dev/MeshCore by GitHub user mikecarper, on the keymindCascade branch. MIT licensed (the copyright line is inherited from upstream, Scott Powell / rippleradios.com). Tracks upstream v1.16.x.
It is development firmware. Every version string carries -dev-, builds are pinned to a commit hash rather than a stable tag, and some releases are flagged pre-release. It moves fast. Treat it accordingly, and do not put it on a node you cannot easily reflash.
It exists to fixThis one thing, and it is worth understanding properly,understanding, because theit fixes a problem it targets is one most people on a multi-hop meshyou have alreadyprobably hit without knowing what caused it.had.
The problem: messages die silently at one bad hopproblem
StockYou MeshCoresend hasa message. Your app says it went. It never arrives.
Here is why. When your message travels through several repeaters to reach someone, no hop-by-hop acknowledgement and no hop-by-hop retry. Aeach repeater forwarding a direct-routed packet transmitssends it onward exactly once and then forgets about it. If the next hoprepeater didmisses not hear that transmissionit (a collision, a moment of interference, thetwo nextradios repeatertalking wasat mid-transmit and therefore deaf, a brief fade)once), the packet is simply gone. No node in the chain knows or cares.
The only retry anywhere in stock MeshCore is end to end: the destination sends an ACK back to the originator, and if the originator does not get that ACK it retries, eventually giving up and falling back to a flood. That has three consequences you have probably experienced:
This is thenot expensive, network-wide fallback. So per-hop unreliability converts directly into mesh-wide airtime.
How bad is it in practice?rare. One upstream user measured it: "anroughly unreliability45% of approximatelydirect 45%messages fromfailed my location as soon asonce the path lengthwas two or more hops. There is greatera thanrelated orbug equalwhere toyour 2.message Thisactually meansdid thatarrive almost half ofbut the direct messages fail via a partially functioning path with more than 2 nodes." Another upstream user, testing a four-node chain over weak links,receipt got 81 of 100 packets deliveredlost on stock.the way back, so your app says it failed and you sit there re-sending.
The fix: listen for the echofix
TheCascade centralmakes ideaeach isrepeater elegantlisten andto costscheck nothingits whenmessage thegot networkpicked is working.up.
AWhen a repeater forwardspasses ayour packet,message along, it then listens for the next hoprepeater to forwardpass it on too. Hearing that same packet. Overhearing that onward retransmission is anproof implicit,it free, hop-local acknowledgement: an echo.worked. If theit echohears arrives,nothing, the queued retry is cancelled and never transmits. If no echo arrives within a computed window, the repeaterit assumes the hopmessage was lostdropped and sends it again.
ThatThink singleof mechanismit turnsas handing something down a fire-and-forgetline forwardingof chainpeople intoand watching to make sure the next person actually takes it, instead of dropping it on the floor and walking away.
The neat part: when everything is working, this costs nothing. No extra messages get sent. The radio only speaks up again when a per-hop-confirmedmessage chain,genuinely withoutgot addinglost. anyIt new packet type and without a single extra transmission indoes the successsame case.for Airtimedelivery is spent only where a hop actually failed. It applies to text messages, path and trace packets, requests and responses, and to ACK packets themselves,receipts, which is what addressesfixes the "it says failed but actuallythey delivered"got bugit" above.problem.
EverythingIt elsealso infixes theone-way forkpaths, iswhere machineryyour around that idea:
recent.repeateroutpathaltpathoutpathaltpathBefore you runflash ait, cartry repeaterthis andinstead
A housegood repeater, your house hearing your car re-flood a packet from three metres away proves nothing about whether the packet escaped the valley. flood.retry.ignore lets you discount it; flood.retry.prefixes lets you define success as "a repeater that actually matters heard it", such as the mountain-top site you need to reach.
Half of itCascade is not new code
Worthat knowingall. before you flash anything: a good part of the Cascade delivery storyIt is stockordinary MeshCore featuressettings that stock ships turned off. CascadeYou justcan turnsswitch them on.on yourself, on official firmware, right now:
path.hash.mode=2 loop.detect=minimal rxdelay=2
agc.reset.interval=8 advert.interval=0 flood.advert.interval=83set multi.acks=1 companion.manual.add=1 companion.autoadd=0path.hash.mode=2multi.acks=1That costs nothing, needs no fork, and losinggets ityou makespart a delivered message look failed, so paying a little airtime to protect it is a reasonable trade.
rxdelay=2loop.detect=minimalagc.reset.interval=8companion.autoadd=0The practical upshot: you can adopt that settings profile on stock firmware right now, without flashing a fork.way. Only the echo-retry,listen-and-retry adaptive coding rate and outpath machinerybehaviour actually requires Cascade.
The catch
Retries use airtime, and airtime is shared. Your retries are everyone else's interference. Cascade ships three settings profiles:
infra
Few
You are in a busy area with lots of nodes. Other routes exist, so do not shout.
rooftop (default)
Many
A fixed node with a weak-ish link into a mesh that mostly works.
mobile
Most
You are out at the edge with no other way through, and getting the message out beats being polite.
The rule: the busier your area, the fewer retries you wantshould mostuse. ofA rooftop repeater in a well-covered city should probably be on infra, not rooftop, whatever the airtimename andsuggests. routingPutting hygienemobile withouton runninga developmentbusy firmware,repeater setwill themake stockthings valuesworse byfor handeveryone andaround stop there.you.
Does it actually work? AnHonestly, honest answerpartly
Cascade's own documentation contains no performance numbers at all. No before-and-after delivery rate, nothing. It is a reference manual for the settings, not a results paper. Do not let anyone tell you it ships with a validated claim.
The evidence that exists is mixed, and it splits cleanly:
- The
conceptidea iswell supported.proven.ASomeonecompeting implementation ofbuilding the sameecho-retryfeatureidea wasseparately tested it on real hardware overa chain ofweak links andtook deliverywent from 81 of 100packetsmessages arriving to 96 of 100.In simulation the same author took a 71% delivery rate on stock to 95.5% with two retries. Hop-by-hop retry with echo cancellation clearly works. - Cascade's
specificowntuningsettingsisarecontested.disputed. The one head-to-headsimulation availabletest saysCascade's aggressiveness works against it. With up to 15Cascade retriespersorepeater,hard that its own retries collide with each other, delivering less than asinglegentlermessageversion.wasTheestimated to generate around 26 direct transmissions, and the resulting collisionsreduceddelivery compared with a simpler implementation using two retries. mikecarper'author's ownsimulator runtesting showed only a+2%delivery improvement, and he acknowledged it "seems to show this helps; better than showing it hurts."improvement.
His defence of the high retry counts is worth quoting, because it is directly relevant to us: "I'm in the PNW and the mesh here is huge. I need more than 3 tries to get out there; sometimes a lot more." His numbers are tuned for a large, congested, long-path Pacific Northwest mesh. That is the mesh Meshwe Americaoperate in.
Also worth knowing: it is buildingexperimental in.firmware (every release is marked as a development build), and MeshCore has not adopted any of this yet, not because they rejected it but because two people built competing versions and nobody has picked one.
The cost, and how to set itrest
RetriesWidest arehardware airtime, and in a shared channel your airtime is somebody else's collision. Three presets ship:
infrarooftopmobileRead that table backwards and it tells you the rule: the denser and busier your RF environment, the fewer retries you should run. In a well-covered mesh your retries are other people's collisions and there are other routes anyway. A rooftop repeater inside a covered metro is arguably an infra node, not a rooftop one, whatever the preset is called. Putting mobile on a busy repeater will degrade the mesh for everyone on it.
Two hard costs to know about:
The known weaknesssupport of the whole approach was named by upstream contributor samm-git when it was first proposed: "This could easily flood the air in noisy areas, where peers hear you but you don't hear the echo." That is exactly right. On an asymmetric link the next hop did get your packet, you just did not hear its echo, so you retransmit something that was already delivered. The recent-repeater signal gate is a mitigation, not a cure.
Why upstream has not merged it
Not because the idea was rejected. As of writing, every relevant pull request is open and unmerged, and dynamic coding rate is on MeshCore's own published roadmap as a to-do. What has happened instead is that upstream has not converged on whose implementation, or on how aggressive it should be. There are now two competing pull requests for the same feature, the Cascade ones are large and carry conflictssix (mikecarper84 callsdevices, it23 hismanufacturers). "kitchenBuilds sink branch"), and one of them changes 74 files and has attracted no review comments at all.
mikecarper's own position is worth repeating: "I don't care how this gets put into the code as long as we have some kind of retry on DM's that's a win." He is probably right, and the thing to watch is the upstream discussion rather than this fork specifically.
Roles, builds and gotchas
Not documented anywhere: what "Halo", "Keymind" and "Cascade" actually refer to. The repository does not say and there is no third-party coverage of this fork. We are not going to guess.configurator.
MCLite
Maintained by GitHub user laserir at github.com/laserir/MCLite. MIT licensed. NotTurns a fork:LilyGo itT-Deck isPlus or T-Watch Ultra into a separate application that pulls MeshCore in as a library and puts its own interface on top. Pre-1.0, currently the 0.4.x series, and described by its own author as experimental software, not a certified radio product.
The point of MCLite is a radiomessenger that works entirely on its own. No phone, no pairing, no account. Turn it on and messagetext people. MIT licensed, still pre-1.0, and its author calls it experimental.
Role:companion only. ItSetup isnotarepeater,fileroomonserver or sensor. It talkstothose.
config.jsonCompatibility, with the caveat. MCLite says it is compatible with other MeshCore devices and the official iOS and Android apps, and it works as a normal companion radio over BLE, Wi-Fi or USB (one transport,The one clientreal atcatch: a time). But "fully compatible" overstates it, and the project's own documentation shows why:
path_hash_modeWADAMESH
MaintainedA byfull ALLFATHERtouchscreen BVinterface: (Belgium)chat, atcontacts, github.com/ALLFATHER-BV/wadamesh,and a real pannable map with aoffline projecttiles, site at wadamesh.com. GPL-3.0-or-later. Like MCLite it is an application rather than a fork: an LVGL touch interface built against an MIT-licensed MeshCore fork the same organisation maintains.
Its own summary: a full touch-screen interface for MeshCore, with the same protocol and the same phone-app companion as stock, just a richer front endall on the device. Works with the phone app at the same time. GPL licensed, from ALLFATHER BV in Belgium.
Role:companion only, with an on-device UI.
Everything is beta. There is no 1.0.0 Releasesrelease areyet.
numbered beta_NN. beta_40 was
The Heltec V4 ishas theless constrained board. 2 MB of PSRAMmemory and no SD card slot, andso several features are reducedcut or absentback on itit.
ZephCore
MaintainedMeshCore byrebuilt GitHub user liquidraver at github.com/liquidraver/ZephCore. MIT licensed. The most architecturallyon different ofunderlying software (the six: a port of MeshCore from Arduino to the Zephyr RTOS (a real-time operating system), aimingwhich atlets fullthe protocolradio compatibilitysleep properly between messages instead of idling. MIT licensed. Aims to be fully compatible with thenormal Arduino firmwareMeshCore and the MeshCore mobilephone apps.
The reason that matters is power. The Arduino build runs a cooperative loop() and the CPU busy-waits unless something explicitly sleeps it. ZephCore is event driven, so the CPU idles between events instead of spinning. On top of that:
LoRaBig battery savings, but you have to switch them on. Its power-saving receivedutymodecycling.cutsThe radio sleeps between short listening windows instead of receiving continuously. The project reports this cutting LoRa receiveidle current from roughly 10 to 15 mA down toroughly3 to 5mA. That is the project's own figure, with no stated method, and we know of no independent measurement.mA,Itbut it is off by default.You turnTurn it onper node at runtimewithset rxduty on.It is not available on LR1110 (a mid-preamble lock issue) or SX127x radios.If youare choosingpicked ZephCore for battery life, this is the switch you camefor,for and it is notthrownflipped for you. (Not available on LR1110 or SX127x radios.)AdaptiveSmartercontention window.timing.StockItMeshCoremeasureshashowthreecrowdedstaticthe airwaves are locally and waits accordingly, instead of using one fixed delaysettings (txdelay,rxdelay,direct.txdelay) that apply the same retransmit jitter regardless of local conditions. ZephCore replaces them with a system that measures actual contention: it counts how many neighbours retransmit the same flood packet it just sent, and sizes its delay from that. Near-zero delay in a quiet linear chain, more in a dense cluster. The project states this is local behaviour with no changes to what goes over the air, so it works alongside Arduino repeaters.everywhere.Adaptive Power Control.Compiled in, off by default. Enable it per node and it reduces transmit power when echo signal-to-noise shows excess margin.
How to choose
DefaultJusttouse MeshCore Official. If nothing below applies, flash the official build and stop reading.If your board has a screen, the official GUI build is included.DirectMessagesmessageskeepfailing on multi-hop paths?failing?That is the problem Keymind Cascade was built for, and no other distribution addresses it. Before you flash a development build,First try the four stock settingsprofileinfromthethatKeymindsectionsection.(multi.acks,Theyrxdelay,areloop.detect,agc.reset.interval), which costs you nothing and needs no fork.free. Ifyouit is stillneed per-hop retries,bad, flash Keymind Cascade and pick thepresetprofile that matches how busy your areaactuallyreallyis, not how the preset is named.is.- Solar or battery
repeater on a tight power budget?repeater? EasySkyMesh PowerSaving or ZephCore.Both attack idle draw, from different angles. Size your battery off the figure for your specific board, not the headline. If you useWith ZephCore, remember toactuallyturn onrxduty. - T-Deck, T-Watch or
anotheratouchscreentouchscreen?board,MCLiteandto use it without a phone. WADAMESH if you wantit usable without a phone?MCLite forthekeyboard and SOS, field-deployment angle. WADAMESH for the map and touch-UI angle. Both are companion-only and both are pre-1.0.map. Your board isBoard notonin the official flasher?CheckTry the Keymind Cascadeandor ZephCore catalogs.Between them they cover a lot of hardware upstream does not build for.
WhateverOne youlast pick,thing. theNone of this matters if your radio settings are wrong. Your node still has to be on the rightsame frequency, bandwidth, spreading factor and coding rate foras everyone else in your regionarea. and your local mesh. Firmware choice does not change that, and a mismatch hereThat is athe verymost common reason a new node hears nothing at all. Get those four valuesnumbers from your local mesh communitygroup before you go lookingblaming forthe firmware bugs.firmware.
A noteword onof maintenancecaution
TheseFour of these are small community projects of varying size. Upstream MeshCore has a large contributor base; ZephCore and WADAMESH have several contributors each; MCLite and EasySkyMesh are veryopenly smallexperimental. teams. Several of these distributions are explicitly beta or pre-release.
That is not a reason to avoid them.them, Itbut it is a reason to keep a stock build handy, to know how to reflash a node you cannot easily reach, and to think hardtwice before putting experimental firmware on a repeater at the top of a tower.
Sources: the Mesh America Device Configurator provider catalogs (apps.meshamerica.com);catalogs; the official MeshCore flasher catalog (flasher.meshcore.io);catalog; and the repositories, release notes, documentation, issues and pull requests of each project. For theThe deliverability section specifically: docs/halo_keymind_settings.md and draws on the keymindCascade docs/cli_commands.mdbranch;branch documentation, MeshCore discussion #1625; MeshCore#1625, pull requests #2367#2367, (HALO direct retry), #2501 (Keymind flood retry),#2501, #2586 (outpath),and #2670 (competing implementation, with the hardware#2670, and simulatorissues delivery#1342, figures)#1775 and #1923 (competing dynamic coding rate); and MeshCore issues #1342 (the 45% multi-hop failure measurement), #1775 (path instability and hash collisions) and #1834 (delivered messages reported as failed).#1834.
Where a figurenumber comes from a project's own testing and nobody has not been independently verified,checked it, this page says so.