Fastlane alternatives in 2026, job by job

"Replace fastlane" is not one decision. Fastlane is eight or nine tools sharing a command, and which alternative you want depends entirely on which lanes you actually run. Nobody replaces all of it with one product.

So start by listing what your Fastfile does. Then swap the parts that hurt and keep the parts that do not.

And before any of that: if the only thing wrong is that nobody has touched the Fastfile in a year, the cheapest fix is usually an afternoon updating gems, not a migration.

Fastlane, broken into its jobs

What you use it for	The action	What replaces it
Build the binary	gym / build_app	Codemagic, Bitrise, Xcode Cloud, EAS Build, plain xcodebuild / Gradle in CI
Code signing across a team	match, sigh, cert	Xcode automatic signing, managed signing in Codemagic or Bitrise, Xcode Cloud
Push to TestFlight	pilot	Xcode Cloud, EAS Submit, Codemagic, Transporter
Upload IPA and metadata to Apple	deliver	App Store Connect API directly, Transporter, EAS Submit plus EAS Metadata, AppSubmit, Lance
Upload AAB and metadata to Play	supply	Play Developer API directly, EAS Submit, Codemagic, Bitrise, AppSubmit
Capture screenshots	snapshot, screengrab	Xcode UI tests run directly, Maestro, or capture by hand
Frame and caption screenshots	frameit	template tools (AppScreens, Previewed and similar), design tools, AppSubmit's studio
Pre-check metadata for rejection risk	precheck	AppSubmit's listing lint, or reading the guidelines yourself
Bump versions, tag, changelog	assorted actions	your CI's own scripting, or a release-management product

Nothing in the right-hand column is strictly better. Some of it is paid, some of it is narrower, and one column entry ("directly") means writing the HTTP calls yourself.

Managed CI that publishes for you

If your reason for leaving is "I do not want to own a Ruby toolchain on a build machine", this is the category to look at. All prices below were read from vendor pricing pages on 17 August 2026.

Codemagic. Free tier of 500 macOS M2 minutes a month with 1 parallel build and 1 team member. Pay as you go at $0.095 per Mac mini M2 minute, $0.114 per M4 minute, $0.045 for Linux and Windows, extra concurrency $49. Fixed annual plans at $3,990 (M2), $5,400 (M4) and $9,000 (M4 Max), each with 3 concurrent builds and unlimited minutes. Enterprise from $12,000 a year. Publishing is declared in the same YAML as the build: IPA to TestFlight or App Store review with phased release, AAB to any Play track with a rollout fraction. Its own docs note that you create the app record and upload the first version manually, and that screenshots, privacy policy URL and category stay in App Store Connect.

Bitrise. Free Hobby tier with 1 private app, 5 concurrencies and 300 credits a month. Starter listed at "$99 per month" billed monthly with up to 3 concurrent builds, 10 team members and 2 private apps. Pro from "$218 per month" with 10 concurrent builds and unlimited team members. Enterprise custom, from 2,000 builds a month. Add-ons priced separately, including Build Cache from $600 a month and Release Management at $250 per app per month on the Standard tier. The step library is the draw: a large catalogue of prebuilt steps, several of which wrap fastlane anyway.

Xcode Cloud. Apple's own, inside Xcode and App Store Connect, with TestFlight built in. 25 compute hours a month come with the Apple Developer Program, then $49.99 for 100 hours, $99.99 for 250, $399.99 for 1,000 and $3,999.99 for 10,000. Signing is managed for you, which removes the single most common reason people needed match. Apple platforms only, so an Android app needs a second answer.

EAS Build and EAS Submit. If you are on Expo, this is the shortest path. EAS Submit uploads a signed AAB to your chosen Play track and a signed IPA to App Store Connect, where it appears in TestFlight after processing. Expo's docs are explicit: it "uploads your binary but does not manage store listing metadata, screenshots, or release notes", though EAS Metadata handles Apple app information and descriptions.

GitHub Actions and friends, without fastlane

You can run xcodebuild and Gradle in any CI and hand the artifacts to the store APIs without fastlane in the loop. On the Apple side that usually means Apple's Transporter (iTMSTransporter) with an App Store Connect API key, plus API calls to create the version, attach the build and submit. On the Play side it is the Play Developer API: open an edit, upload the bundle, assign a track, commit.

This is real work, but it is finite and it is yours. The App Store Connect API and the Play Developer API are both documented and stable, and once you have written the version-and-submit chain once it rarely changes.

Two cautions. First, macOS runners are the expensive part of any hosted CI, which is why the per-minute rates above matter more than the plan names. Second, xcrun altool has been progressively replaced for upload purposes and Apple points at Transporter; check the current guidance rather than copying a 2021 blog post.

Screenshots without snapshot

snapshot is the lane people abandon most often, because it means writing UI tests whose only job is producing marketing images, and those tests break whenever the UI moves.

The alternatives, honestly:

1. Keep the UI tests but run them directly. xcodebuild test with screenshot attachments, no fastlane, if the tests already exist for real testing reasons.
2. Maestro. A cross-platform mobile UI automation tool that is considerably less painful to write than XCUITest, and can drive a capture flow.
3. Capture by hand, then automate the rest. Six screens in a simulator takes ten minutes. It is the resizing to every store slot and the framing that is worth automating, not the pressing of the shutter.
4. Template tools. AppScreens, Previewed and similar products frame and caption screenshots from templates.

Option 3 is the one most solo developers land on eventually, and it is not a defeat.

Store-facing tools, if that is the half you want gone

Fastlane's deliver, supply and precheck are the store-facing lanes, and there is now a small category of hosted products aimed at exactly them.

AppSubmit (this site) takes a signed IPA or AAB and handles the rest: validating screenshots against every slot each store accepts, resizing without stretching, AI listing copy with a rejection-risk lint, delivery to App Store Connect via Transporter and to a Play track via the Edits API, creating the Apple version and submitting for review, then decoding rejections and drafting review replies. Free tier, Indie $29/month, Studio $79/month. It does not build and does not sign, which is the honest reason it is not a full fastlane replacement. Private beta with a waitlist as of August 2026.

Lance covers more of the chain, including building and signing on its own infrastructure, driven through an MCP server in Claude Code, Cursor or Codex. Free starter credits for 30 days, then $40 a month for 2,000 credits, overage $20 per 2,000. Backed by Y Combinator. Its material describes App Store work throughout.

appsubmit.ai (a different product, similar name) generates listing metadata and assets with GitHub repository import: free tier, $19 a month, $49 a month.

appsubmitter.io is a person doing it for you: $289 iOS, $259 Android, $419 both, per submission, at prices shown on 17 August 2026.

Or nothing at all

App Store Connect and Play Console are free with your accounts (99 USD a year for Apple, a US$25 one-time fee for Google), they are authoritative, and they are always current with themselves.

If you built a fastlane setup for one app that ships three times a year, deleting it and using the consoles is a legitimate outcome. The automation was costing more than the work.

Pick by why you are leaving

1. "I cannot maintain the Ruby toolchain." Managed CI: Codemagic, Bitrise, or Xcode Cloud if you are Apple only.
2. "Code signing keeps breaking." Xcode Cloud or managed signing in Codemagic or Bitrise. This alone removes match.
3. "I am on Expo." EAS Build plus EAS Submit. Do not fight it.
4. "I want fewer dependencies, not different ones." Write directly against the App Store Connect API and Play Developer API in your existing CI.
5. "Screenshots are the worst part." Drop snapshot, capture manually or with Maestro, and automate resizing and framing instead.
6. "The store paperwork is what eats my week." A store-facing tool: AppSubmit, Lance, or appsubmit.ai depending on scope.
7. "I ship rarely and I am tired." The consoles.
8. "Actually nothing is wrong, it is just old." Update the gems. Keep fastlane.

FAQ

Is fastlane deprecated? No. It is active, MIT licensed, around 42,000 stars, with a large maintainer list. People leave because of the maintenance burden on their own setup, not because the project stopped.

What is the closest single replacement for fastlane? There is not one. The nearest thing is a managed CI product with publishing built in (Codemagic, Bitrise), and even then screenshots and listing metadata land somewhere else.

Can I drop fastlane but keep match? Yes. match is usable on its own, and plenty of teams keep it purely for certificate and profile sync while moving building and publishing elsewhere. It is often the last lane standing.

Do I need fastlane to submit to Google Play? No. The Play Developer API does it in four calls: open an edit, upload the bundle, assign it to a track with release notes, commit the edit.

Will any of these stop rejections? No. They reduce the avoidable ones, such as wrong screenshot dimensions or beta language in a description. A reviewer's judgement is not automatable by anyone.

Is there anything fastlane does that nothing else does? Keeping localised store metadata as reviewable files in your repository, cleanly, for free. If that matters to you, it is a strong reason to stay.