Should you evaluate this beta?
Evaluate v2.0.0-beta.25 with separate data if its shared desktop and headless design fits your workflow. Motrix v2.0.0-beta.25 warns against using your only copy of v1 data. Its release notes say migration had not been validated and recommend a separate OS account, machine or Docker data directory. They also warn against relying on the beta for your only copy of important downloads.
Those notes describe downloads as planned after release gates pass. They do not establish that every planned package was published. Check availability before planning an evaluation around a desktop package or container image.
This is a dated beta evaluation. The supplied current README describes Motrix 2.2.0, a Migration Assistant and remote browser-extension pairing with Server. Those later documented capabilities do not establish behavior at our tested revision, and the beta's restrictions should not be treated as permanent product limitations.
What our run established
Our run began on 2026-08-24 at 19:24:36 UTC, using commit b27f90a (full SHA b27f90aaa7463c983260d9e8c16c78b7c4fc9fff). The sandbox had 3 CPUs, 8 GB of RAM and the lab-node:22 image. The install log identifies Node 22.23.2.
Our sandbox installation succeeded in 27 seconds. Installation returned exit code 0; the build returned exit code 1 after 27 seconds, and the test command returned exit code 1 after 386 seconds. Neither failed command reached the harness timeout. Completing installation does not establish that the application could download files or run a desktop session.
Why the build stopped
The native-host build stopped when Rust could not find std for x86_64-unknown-linux-musl. The log says the target may not be installed and suggests rustup target add x86_64-unknown-linux-musl. Cargo returned 101, causing the build command to fail. We did not apply that suggestion or rerun the build, so it remains an unverified remedy.
Why three tests failed
The server-package smoke tests required Node 24, while our sandbox ran Node 22.23.2. In tests/scripts/smoke-server-package.test.ts, the orchestration test stopped at that runtime check. The missing-aria2 test encountered the same check before reaching its expected missing-file error. These failures therefore do not establish successful server startup or missing-binary handling.
Separately, src/server/plugin/install-service.integration.test.ts timed out after 5,000 milliseconds during its installation lifecycle test. The supplied log does not establish the cause. It also prints react-i18next warnings without identifying them as the cause of that timeout.
The stored parser summaries report 7,261 passed, 3 failed and 4 skipped, with a total of 7,268. They do not establish exhaustive unique test coverage: no complete-log reconciliation or new execution accompanied this correction. Repeated summary lines are not additional executions, and skipped tests are not failures. The harness recorded 386 seconds for the command; Vitest displayed 353.29 seconds. These are different timing scopes.
The recorded footprint is 581 MB. The stored dependency counter is 1, but the install log shows hundreds of packages being added, so that counter is not a complete dependency inventory. No audit result or scanned-package coverage is recorded.
We did not validate HTTP or torrent transfers, transfer speed, desktop controls, browser pairing, CLI operation, Docker deployment or migration. This correction includes no new execution. The MrKeyoor testing-methodology page explains the reporting approach and its limits; its current text does not establish which methodology revision governed this August run.
What the tested revision documents
The README at our exact commit documents HTTP, FTP, BitTorrent and magnet support, SQLite-backed session recovery, and a shared core for desktop and headless operation. These are documented capabilities, rather than workflows verified by our run.
Browser extensions and command-line clients communicate through MDXP (Motrix Download eXchange Protocol), which the README describes as an open protocol built on JSON-RPC 2.0. The revision also documents a CLI, browser extensions and a plugin toolchain. Their presence establishes the documented ecosystem's scope; it does not measure community participation or support quality.
Plugins run in QuickJS without Node.js APIs or direct file and network access. They declare capabilities and URL-scoped host permissions that Motrix presents to the user before granting access. This describes an isolation design. Our run was not a security assessment.
For a NAS evaluation, distinguish application state from downloaded files. The supplied Docker deployment guide documents separate writable /data and /downloads mounts. The exact-commit README requires MOTRIX_PUBLIC_URL to identify a Web approval URL that remote clients can reach, and requires a TLS reverse proxy and firewall rules for internet or untrusted-LAN access. We did not deploy this configuration. The current deployment guide contains later features, so use documentation matching the version you choose.
At the tested commit, the README says first-time browser-extension pairing is provided by the desktop app and not the headless server. The current README documents remote Server extension pairing. Keep that version distinction when deciding whether a NAS-only installation meets your needs.
When another tool fits
aria2 supports HTTP(S), FTP, SFTP, BitTorrent and Metalink downloads. Its README documents a command-line interface, daemon operation and RPC interfaces. Consider it as a workflow replacement when those interfaces meet your needs without Motrix's application interface. Motrix documents a maintained aria2 fork, so do not assume upstream aria2 can replace its bundled engine internally.
qBittorrent is a BitTorrent client built with C++ and Qt using libtorrent. Its repository supports considering it for a dedicated BitTorrent workflow. The supplied extract does not support a claim that its torrent controls are superior to Motrix's.
Neither alternative was tested as a companion to Motrix in this run.
What the ratings assess
The numerical ratings are editorial judgments, not performance measurements. For assessed dimensions, the scale runs from 1 for major documented obstacles, through 2 for substantial friction, 3 for a documented approach with material conditions, and 4 for broad supporting evidence with remaining limits, to 5 for comprehensive documentation of the reviewed workflows. A score does not establish adoption or production reliability. Unassessed dimensions are shown as N/A and excluded from numerical aggregates.
Setup ease is 3: the documented installation paths coexist with an unresolved source-build prerequisite and a smoke-test runtime mismatch. Packaged installation was not tested.
Documentation is 5 for the supplied material's coverage of desktop installation, headless storage and networking, plugin permissions and development. This judges documented coverage, not successful execution of those workflows. The exact-commit README's Node 22 minimum did not explain the Node 24 smoke requirement encountered in our run.
Community is unassessed. The supplied evidence documents integrations and contribution routes but does not measure contributor participation, user adoption or support response quality.
Maturity is 3 for this prerelease revision: it documents an isolated evaluation path and explicitly warns that v1 migration had not been validated. Our run does not establish operational reliability.

