I want to choose between these options I do not care about immaturity, instead I see that as potential

Grok 4.3

Vote: Xilem

Given your explicit focus on potential over maturity, "Xilem" stands out as the forward-looking choice among the options. It represents an evolving Rust UI approach with strong architectural ideas that align with long-term growth.

Claude Opus 4.7

Vote: Xilem

Given that the user explicitly states they do not care about immaturity and instead see it as potential, this points strongly toward the most experimental and forward-looking option. "Ratatui" is a mature, stable TUI library — it doesn't fit the 'immaturity as potential' framing. "Dioxus Native" and "GPUI" are interesting but more established in their trajectories (Dioxus is a well-known React-like framework, GPUI powers Zed and is relatively proven). "Xilem" is the most architecturally ambitious and experimental of the group — Raph Levien's reactive Rust UI framework built on a novel view tree / ECS-inspired architecture. It is famously immature/in-flux but has enormous potential as a next-generation Rust GUI foundation. This matches the user's stated preference perfectly.

Gemini 3.5 Flash

Vote: Xilem

Since you view immaturity as potential, **Xilem** is the most compelling choice. Led by Raph Levien and the Linebender community, Xilem is an experimental, next-generation UI framework exploring a reactive, architecture-first approach (inspired by SwiftUI and React) tailored specifically for Rust's ownership model. It leverages **Vello**, a highly advanced, GPU-accelerated rendering engine utilizing compute shaders. While currently highly immature and not yet ready for production, its architectural elegance and cutting-edge rendering pipeline give it the highest theoretical potential to solve the long-standing "Rust GUI problem" in the long run. If you want to be at the absolute bleeding edge of Rust UI research, Xilem is the project to watch and contribute to.

GLM 5.1

Vote: Xilem

The phrase 'I do not care about immaturity, instead I see that as potential' strongly aligns with Xilem (option D). Xilem is the newest and most experimental framework among the options, being developed by the Linebender team as a complete architectural rethink of how UI frameworks should work in Rust. It's built on a data-oriented design philosophy and learns from the shortcomings of Druid. Its immaturity means there's vast potential for contributors to shape its direction and be early adopters. If you see immaturity as potential rather than a drawback, Xilem's early stage represents the most opportunity for growth and impact, making it the most fitting choice for your stated preferences.

GPT-5.5

Vote: GPUI

I would choose "GPUI" because it has the best mix of high upside and real-world validation. It is still young, but it is not just theoretical: it powers Zed, which proves the core ideas can support a serious, high-performance desktop application. If you see immaturity as potential, "GPUI" is especially attractive because its API and ecosystem can still evolve, while its direction is already clear: fast, native-feeling, GPU-accelerated desktop UI in Rust. "Xilem" is also promising, but it feels more experimental and architecture-driven. "Dioxus Native" has broader app-framework ambitions, and "Ratatui" is excellent if you specifically want terminal UI, but for maximum future potential in native desktop UI, I would bet on "GPUI".

Kimi K2.6

Vote: Xilem

You explicitly embrace immaturity as a sign of future potential rather than a liability. Among the choices, Xilem is the most forward-looking and experimental framework. It is a ground-up architectural rethink of native Rust UI by Raph Levien, built around novel ideas like the Masonry layout system and efficient reactive architectures. It is explicitly positioned as experimental and not yet production-ready. Ratatui is already mature and stable in its TUI niche, while Dioxus Native and GPUI are much closer to shipping real applications today. Because Xilem is currently the most immature yet architecturally ambitious option, it offers the greatest room to grow alongside your own project. Choosing it means betting on the underlying vision and helping shape the ecosystem as it matures.

Final Summary

The AI Roundtable overwhelmingly identified Xilem as the ideal choice for a user who equates immaturity with potential, with five models backing the experimental framework. Claude Opus 4.7 and Gemini 3.5 Flash led the charge, arguing that Xilem’s radical architectural rethink represents the highest theoretical ceiling for Rust UI development. The consensus was nearly total until GPT-5.5 broke ranks, casting the lone vote for GPUI by arguing that true potential is best realized when experimental ideas are backed by real-world performance like that of the Zed editor.

No agreement after 1 rounds of discussion. Final vote: Xilem: 5, GPUI: 1

Strongest Arguments

• GPUI: GPUI offers the best mix of high upside and real-world validation; it is young enough to evolve but proven enough to power high-performance applications like Zed.
• Xilem: As a ground-up architectural rethink built on a novel reactive view tree and GPU-accelerated rendering, Xilem is the most ambitious and experimental project, offering the greatest room for growth and influence.