AvaloniaUI/Avalonia

Choose this fork only if its local DataGrid/navigation tradeoffs are exactly what you need and you are prepared to own long-term maintenance. For most adopters, upstream Avalonia is the safer default because this fork is stale, highly divergent, and appears to remove core navigation functionality.

Choose this fork only if you need the old Perspex lineage or are doing framework archaeology. For new application or platform work, upstream Avalonia is the clearly better choice because this fork is stale and materially behind.

Choose upstream unless you need this exact older state. This fork appears to add no new capabilities and is 187 commits behind, so it is a weak choice for new adoption or ongoing development.

Prefer this fork only if its control/navigation changes or JetBrains-specific workflow support are directly valuable to you. If you want a broadly current, low-risk Avalonia base, upstream is the safer choice because this fork is far behind and significantly diverged.

Choose this fork only if you need its custom framework changes and can own a large divergence burden. For most adopters, upstream is the safer default because it is far more current and actively maintained.

Choose this fork only if you need its specific local changes and are prepared to maintain a highly divergent codebase. For most adopters, upstream Avalonia is the safer default because this fork is stale and materially behind current framework work.

Choose upstream unless you specifically need this fork's custom behaviors and are prepared to own a large, stale divergence. For new adopters, the maintenance risk outweighs the apparent benefits.

Adopt only if you specifically need this fork’s custom behavior and are willing to own a large, outdated divergence. For most users, upstream Avalonia is the safer choice because this fork is stale, far behind, and likely missing years of fixes and platform support.

Choose upstream unless you specifically need this fork as an internal snapshot. It offers no additional capabilities in the supplied evidence and is behind on recent upstream fixes and improvements, so it is a poorer default for new adopters.