WoW: Forever launches on November 4. Its beta runs through October 21 — and access is effectively paywalled, tied to the Skyborne Epic Pack, the Forever Collection or the Collector's Edition, with only a limited free opt-in pool. Into that gap, predictably, steps the oldest institution in this scene: people are running self-hosted Forever realms on their own hardware, based on a TrinityCore fork that targets the beta client itself.

Original editorial illustration by RealmList, generated for this article. This is artwork, not game footage or official promotional material.

What is actually reported

The picture below comes from Warperia's write-up as summarized by Indie Kings — a summary of reports and project claims, not independently verified testing. With that caveat stated once and loudly, here is the landscape as reported:

For the TrinityCore fork, the reported setup paths for Windows are a prebuilt repack (extract, setup, start, launch through a dedicated launcher) and a set of host/guest helper scripts for letting friends join a local realm.

Why this happens now

The mechanics are familiar. Beta access concentrates impressions in a small paid group and leaves everyone else curious; emulation communities have always filled that gap by making the client runnable on community code. What is unusual here is the speed and the target: Forever is not a 2004 client that emulators have had two decades to master. It runs on the same modern code family as retail — Blizzard itself has said Forever inherits Midnight's addon restrictions for exactly that reason — so a TrinityCore fork backing a VMaNGOS-converted world onto the 1.60.1 client is working against a moving, modern target.

That target also moves fast. Build 1.60.1.70338 is one of a string of beta builds we've tracked this month — 69893, 70009, 70235, 70291 — each shifting data layouts. A fork pinned to one build ages quickly, and the reported claim set reflects it: the fork reportedly starts, accepts connections and runs older zones, while how much of the actual new content (the quests, dungeons, Skyborne systems) functions is unknown. Startup success and content completeness are different questions, and nothing public answers the second one.

The part that should make players pause

Reading about this is fine. Joining one is a different calculus:

• Legal exposure has never been clearer. 2026 is the year Blizzard showed it will litigate: Turtle WoW and Stormforge were shut down by agreement after a ruling in Blizzard's favor, Ascension closed by arrangement, and the Turtle injunction reportedly covers assisting others to run similar projects. A public server emulating a beta client Blizzard is actively selling access to is about as loud a target as this scene can produce.
• Security is on the user. Self-described repacks and community cores are exactly the download category covered in our client-checklist guide. Nothing in this scene has a reputation to protect yet.
• The claims are self-reported. "500+ bugs fixed," "all Vanilla content works," "dungeon bots capped" — every one of these is a project describing itself, relayed through a tracker, with no independent testing behind it. The same skepticism we apply to population claims applies double here.

Why it still matters

Even for players who will never run one, the scene's existence says something about the pre-launch moment. Forever was announced as the answer to the Classic+ demand that private servers spent a decade proving; within four weeks of its beta opening, the emulation community is already trying to reproduce it — the same way it once reproduced Classic before Classic existed. Whether these projects remain toys, become the next takedown headlines, or simply evaporate when the official launch arrives on November 4 is worth watching. We won't be listing any of them in the directory until any of the above is independently verifiable — and even then, the legal bar for a Forever-clone is a different category entirely from a 3.3.5a realm.