Why this category
Smart litter boxes and smart pet feeders are among the faster-growing corners of consumer hardware, and several leading brands in this space sell into both the Chinese domestic market and English-speaking markets at the same time. That overlap makes it a natural place to check something our own measurement is built to catch: does an AI model's answer to "what's the most popular brand" change depending on what language you ask it in?
What we measured
Asking "Most Popular Smart Cat Litter Box Brands" in English (measured 2026-09-02 13:01:51, panel completeness 92.5%, 7 head models + 5 long-tail models), Litter-Robot topped the field with an index score of 99.63, with PETKIT (小佩) second at 88.12:
| Rank | Brand | Index |
|---|---|---|
| 1 | Litter-Robot | 99.63 |
| 2 | PETKIT | 88.12 |
| 3 | CATLINK | 60.96 |
| 4 | PetSafe | 53.24 |
| 5 | Neakasa | 24.81 |
Ask the equivalent question in Chinese instead (最受欢迎的智能猫砂盆品牌, measured 2026-09-02 12:58:38, panel completeness 100%), and the order flips: PETKIT leads at 86.34, Litter-Robot drops to second at 70.38 — roughly a 23% gap, the other direction.
The same pattern shows up in a second category measured the same day. Asking about smart pet feeders in English (2026-09-02 15:00:15, panel completeness 100%), PetLibro leads at 88.27 while PETKIT sits fourth at 68.08. Asking the Chinese-language version (2026-09-02 12:53:16, panel completeness 100%), PETKIT leads at 84.53, more than 31% ahead of PetLibro's 64.54.
Three observations
The language split isn't a one-off
Two different product categories, four independent measurements, and the same direction of disagreement shows up twice: PETKIT tops the Chinese-language version of both categories and drops several places in the English-language version of both. Each category has only been measured once so far, so the sample is still small — but the same pattern repeating twice is worth noting rather than dismissing as noise.
The Chinese-language consensus runs deeper
In the litter-box category, the English-language leader's index score is only about 12% higher than the runner-up's; in the same category's Chinese-language version, that gap is roughly 23%. The feeder category shows a similarly wide Chinese-language gap (over 30%). Whatever is driving the split, it isn't just a reordering — how strongly the panel agrees on a leader also differs by language.
Individual models don't all agree with the aggregate
In the Chinese-language litter-box measurement, the seven head models' own top-5 picks aren't uniform: Qwen's top pick was Toletu, a brand that doesn't even make the merged top 10; Gemini's top 5 included a long-tail Chinese-market name ("糯雪") that appears nowhere else in the rankings. The aggregate leader is a weighted composite across the panel, not something every model independently agrees on.
Limits
The Chinese-language litter-box measurement this piece is anchored to had a panel completeness of 100% (7 head models + 5 long-tail models; weighting sources are on the methodology page) and a normal normalization status. The English-language litter-box measurement cited above had a panel completeness of 92.5% — the same limitation applies there. It's also worth noting that our own evaluation panel is itself made up of large language models, whose training data may include the results of other AI rankings or reviews — a possible "self-referential contamination" where a model's impression of a brand partly reflects that the brand was already covered elsewhere, not an independent judgment. This is a known limitation of the current methodology, not something specific to this measurement.
Full model panel, weighting sources, and the scoring formula are on the methodology page.