AI Visibility Monitoring · Sample reports
Two sample reports
Both samples are generated from real data — not screenshots, and no number here is made up. No email required.
Sample 1
Rank monitoring
Generated on the spot from our published weekly data. 6 weekly questions (pet products, each asked once in Chinese and once in English), with each model asked once per period.
Data through 2026-09-27 · 5 periods · 12-model panel
Rank and model coverage, question by question
- PETKIT
- PETLIBRO
- Xiaomi
- CATLINK
- PetSafe
- 95% interval
- Vertical line = panel or methodology change; do not compare across it
Data: public ranking runs #17, #44, #134, #140, #153 · check at /rankings/15
Based on answers from 12 models to one question (asked in Chinese), each model asked once per period. Collected 2026-08-24 to 2026-09-27, 5 periods; 11–12 models returned valid answers per period. Intervals are 95% Wilson intervals. Per-period index scores and the question text are at /rankings/15. A change took effect on 2026-09-27 (seat model swapped · scoring or methodology change); later periods are not compared directly with earlier ones.
Table view
| Brand | 08-24 #17 | 09-02 #44 | 09-13 #134 | 09-20 #140 | 09-27 #153 | Adjacent-period verdicts |
|---|---|---|---|---|---|---|
| PETKIT | #1 · 11/11 · 74%–100% | #1 · 11/12 · 65%–99% | #1 · 11/11 · 74%–100% | #1 · 11/11 · 74%–100% | #1 · 12/12 · 76%–100% | 09-02→09-13 no significant change (z=0.98); 09-13→09-20 no significant change (z=0.00) |
| PETLIBRO | #3 · 9/11 · 52%–95% | #2 · 8/12 · 39%–86% | #2 · 9/11 · 52%–95% | #2 · 10/11 · 62%–98% | #2 · 9/12 · 47%–91% | 09-02→09-13 no significant change (z=0.83); 09-13→09-20 no significant change (z=0.62) |
| Xiaomi | #2 · 10/11 · 62%–98% | #3 · 10/12 · 55%–95% | #3 · 9/11 · 52%–95% | #7 · 7/11 · 35%–85% | #3 · 10/12 · 55%–95% | 09-02→09-13 no significant change (z=-0.10); 09-13→09-20 no significant change (z=-0.96) |
| CATLINK | #4 · 8/11 · 43%–90% | #8 · 4/12 · 14%–61% | #6 · 6/11 · 28%–79% | #6 · 5/11 · 21%–72% | #4 · 9/12 · 47%–91% | 09-02→09-13 no significant change (z=1.03); 09-13→09-20 no significant change (z=-0.43) |
| PetSafe | #5 · 7/11 · 35%–85% | #4 · 10/12 · 55%–95% | #4 · 9/11 · 52%–95% | #4 · 9/11 · 52%–95% | #5 · 9/12 · 47%–91% | 09-02→09-13 no significant change (z=-0.10); 09-13→09-20 no significant change (z=0.00) |
- PETLIBRO
- PetSafe
- WOPET
- PETKIT
- Sure Petcare
- 95% interval
- Vertical line = panel or methodology change; do not compare across it
Data: public ranking runs #50, #135, #141, #154 · check at /rankings/42
Based on answers from 12 models to one question (asked in English), each model asked once per period. Collected 2026-09-02 to 2026-09-27, 4 periods; 10–12 models returned valid answers per period. Intervals are 95% Wilson intervals. Per-period index scores and the question text are at /rankings/42. A change took effect on 2026-09-27 (seat model swapped · scoring or methodology change); later periods are not compared directly with earlier ones.
Table view
| Brand | 09-02 #50 | 09-13 #135 | 09-20 #141 | 09-27 #154 | Adjacent-period verdicts |
|---|---|---|---|---|---|
| PETLIBRO | #1 · 11/12 · 65%–99% | #1 · 11/11 · 74%–100% | #2 · 9/10 · 60%–98% | #1 · 12/12 · 76%–100% | 09-02→09-13 no significant change (z=0.98); 09-13→09-20 no significant change (z=-1.07) |
| PetSafe | #2 · 11/12 · 65%–99% | #3 · 11/11 · 74%–100% | #1 · 10/10 · 72%–100% | #2 · 12/12 · 76%–100% | 09-02→09-13 no significant change (z=0.98); 09-13→09-20 no significant change (z=0.00) |
| WOPET | #3 · 12/12 · 76%–100% | #4 · 10/11 · 62%–98% | #3 · 10/10 · 72%–100% | #3 · 11/12 · 65%–99% | 09-02→09-13 no significant change (z=-1.07); 09-13→09-20 no significant change (z=0.98) |
| PETKIT | #4 · 11/12 · 65%–99% | #2 · 10/11 · 62%–98% | #4 · 9/10 · 60%–98% | #4 · 10/12 · 55%–95% | 09-02→09-13 no significant change (z=-0.06); 09-13→09-20 no significant change (z=-0.07) |
| Sure Petcare | #5 · 11/12 · 65%–99% | #5 · 11/11 · 74%–100% | #5 · 9/10 · 60%–98% | #5 · 10/12 · 55%–95% | 09-02→09-13 no significant change (z=0.98); 09-13→09-20 no significant change (z=-1.07) |
- PETKIT
- Litter-Robot
- CATLINK
- PetSafe
- Homerunpet
- 95% interval
- Vertical line = panel or methodology change; do not compare across it
Data: public ranking runs #46, #136, #142, #155 · check at /rankings/43
Based on answers from 12 models to one question (asked in Chinese), each model asked once per period. Collected 2026-09-02 to 2026-09-27, 4 periods; 11–12 models returned valid answers per period. Intervals are 95% Wilson intervals. Per-period index scores and the question text are at /rankings/43. A change took effect on 2026-09-27 (seat model swapped · scoring or methodology change); later periods are not compared directly with earlier ones.
Table view
| Brand | 09-02 #46 | 09-13 #136 | 09-20 #142 | 09-27 #155 | Adjacent-period verdicts |
|---|---|---|---|---|---|
| PETKIT | #1 · 12/12 · 76%–100% | #1 · 11/11 · 74%–100% | #1 · 11/11 · 74%–100% | #1 · 11/12 · 65%–99% | 09-02→09-13 no significant change (z=0.00); 09-13→09-20 no significant change (z=0.00) |
| Litter-Robot | #2 · 11/12 · 65%–99% | #2 · 11/11 · 74%–100% | #3 · 10/11 · 62%–98% | #2 · 11/12 · 65%–99% | 09-02→09-13 no significant change (z=0.98); 09-13→09-20 no significant change (z=-1.02) |
| CATLINK | #3 · 10/12 · 55%–95% | #3 · 10/11 · 62%–98% | #2 · 11/11 · 74%–100% | #3 · 12/12 · 76%–100% | 09-02→09-13 no significant change (z=0.54); 09-13→09-20 no significant change (z=1.02) |
| PetSafe | #4 · 9/12 · 47%–91% | #6 · 8/11 · 43%–90% | #4 · 9/11 · 52%–95% | #4 · 11/12 · 65%–99% | 09-02→09-13 no significant change (z=-0.12); 09-13→09-20 no significant change (z=0.51) |
| Homerunpet | #6 · 6/12 · 25%–75% | #5 · 7/11 · 35%–85% | #7 · 6/11 · 28%–79% | #5 · 9/12 · 47%–91% | 09-02→09-13 no significant change (z=0.66); 09-13→09-20 no significant change (z=-0.43) |
- Litter-Robot
- PETKIT
- CATLINK
- PetSafe
- Neakasa
- 95% interval
- Vertical line = panel or methodology change; do not compare across it
Data: public ranking runs #47, #137, #143, #156 · check at /rankings/44
Based on answers from 12 models to one question (asked in English), each model asked once per period. Collected 2026-09-02 to 2026-09-27, 4 periods; 11–12 models returned valid answers per period. Intervals are 95% Wilson intervals. Per-period index scores and the question text are at /rankings/44. A change took effect on 2026-09-27 (seat model swapped · scoring or methodology change); later periods are not compared directly with earlier ones.
Table view
| Brand | 09-02 #47 | 09-13 #137 | 09-20 #143 | 09-27 #156 | Adjacent-period verdicts |
|---|---|---|---|---|---|
| Litter-Robot | #1 · 11/11 · 74%–100% | #1 · 11/11 · 74%–100% | #1 · 10/11 · 62%–98% | #1 · 12/12 · 76%–100% | 09-02→09-13 no significant change (z=0.00); 09-13→09-20 no significant change (z=-1.02) |
| PETKIT | #2 · 11/11 · 74%–100% | #2 · 11/11 · 74%–100% | #2 · 10/11 · 62%–98% | #2 · 12/12 · 76%–100% | 09-02→09-13 no significant change (z=0.00); 09-13→09-20 no significant change (z=-1.02) |
| CATLINK | #3 · 10/11 · 62%–98% | #3 · 9/11 · 52%–95% | #3 · 8/11 · 43%–90% | #3 · 12/12 · 76%–100% | 09-02→09-13 no significant change (z=-0.62); 09-13→09-20 no significant change (z=-0.51) |
| PetSafe | #4 · 10/11 · 62%–98% | #4 · 8/11 · 43%–90% | #5 · 9/11 · 52%–95% | #4 · 10/12 · 55%–95% | 09-02→09-13 no significant change (z=-1.11); 09-13→09-20 no significant change (z=0.51) |
| Neakasa | #5 · 4/11 · 15%–65% | #5 · 7/11 · 35%–85% | #4 · 6/11 · 28%–79% | #5 · 5/12 · 19%–68% | 09-02→09-13 no significant change (z=1.28); 09-13→09-20 no significant change (z=-0.43) |
- PETKIT
- Homerunpet
- Shernbao
- 欧科达
- 简斯
- 95% interval
- Vertical line = panel or methodology change; do not compare across it
Data: public ranking runs #51, #138, #144, #157 · check at /rankings/45
Based on answers from 12 models to one question (asked in Chinese), each model asked once per period. Collected 2026-09-02 to 2026-09-27, 4 periods; 10–12 models returned valid answers per period. Intervals are 95% Wilson intervals. Per-period index scores and the question text are at /rankings/45. A change took effect on 2026-09-27 (seat model swapped · scoring or methodology change); later periods are not compared directly with earlier ones.
Table view
| Brand | 09-02 #51 | 09-13 #138 | 09-20 #144 | 09-27 #157 | Adjacent-period verdicts |
|---|---|---|---|---|---|
| PETKIT | #1 · 11/12 · 65%–99% | #1 · 8/10 · 49%–94% | #1 · 10/11 · 62%–98% | #1 · 12/12 · 76%–100% | 09-02→09-13 no significant change (z=-0.79); 09-13→09-20 no significant change (z=0.71) |
| Homerunpet | #2 · 9/12 · 47%–91% | #2 · 7/10 · 40%–89% | #2 · 7/11 · 35%–85% | #2 · 9/12 · 47%–91% | 09-02→09-13 no significant change (z=-0.26); 09-13→09-20 no significant change (z=-0.31) |
| Shernbao | #4 · 4/12 · 14%–61% | #3 · 3/10 · 11%–60% | #77 · 1/11 · 2%–38% | #3 · 6/12 · 25%–75% | 09-02→09-13 no significant change (z=-0.17); 09-13→09-20 no significant change (z=-1.22) |
| 欧科达 | #6 · 3/12 · 9%–53% | — · 0/10 · 0%–28% | — · 0/11 · 0%–26% | #4 · 1/12 · 1%–35% | 09-02→09-13 no significant change (z=-1.70); 09-13→09-20 no significant change (z=0.00) |
| 简斯 | #41 · 1/12 · 1%–35% | — · 0/10 · 0%–28% | #80 · 1/11 · 2%–38% | #5 · 2/12 · 5%–45% | 09-02→09-13 no significant change (z=-0.93); 09-13→09-20 no significant change (z=0.98) |
- PETKIT
- Bissell
- Homerunpet
- SHELANDY
- Flying Pig Grooming
- 95% interval
- Vertical line = panel or methodology change; do not compare across it
Data: public ranking runs #49, #139, #145, #158 · check at /rankings/46
Based on answers from 12 models to one question (asked in English), each model asked once per period. Collected 2026-09-02 to 2026-09-27, 4 periods; 10–12 models returned valid answers per period. Intervals are 95% Wilson intervals. Per-period index scores and the question text are at /rankings/46. A change took effect on 2026-09-27 (seat model swapped · scoring or methodology change); later periods are not compared directly with earlier ones.
Table view
| Brand | 09-02 #49 | 09-13 #139 | 09-20 #145 | 09-27 #158 | Adjacent-period verdicts |
|---|---|---|---|---|---|
| PETKIT | #1 · 5/10 · 24%–76% | #1 · 8/11 · 43%–90% | #2 · 5/11 · 21%–72% | #1 · 6/12 · 25%–75% | 09-02→09-13 no significant change (z=1.07); 09-13→09-20 no significant change (z=-1.30) |
| Bissell | #3 · 2/10 · 6%–51% | #4 · 5/11 · 21%–72% | #12 · 2/11 · 5%–48% | #2 · 4/12 · 14%–61% | 09-02→09-13 no significant change (z=1.24); 09-13→09-20 no significant change (z=-1.37) |
| Homerunpet | #5 · 4/10 · 17%–69% | #3 · 4/11 · 15%–65% | #1 · 5/11 · 21%–72% | #3 · 2/12 · 5%–45% | 09-02→09-13 no significant change (z=-0.17); 09-13→09-20 no significant change (z=0.43) |
| SHELANDY | #4 · 2/10 · 6%–51% | #8 · 3/11 · 10%–57% | #8 · 2/11 · 5%–48% | #4 · 4/12 · 14%–61% | 09-02→09-13 no significant change (z=0.39); 09-13→09-20 no significant change (z=-0.51) |
| Flying Pig Grooming | #2 · 7/10 · 40%–89% | #2 · 7/11 · 35%–85% | #3 · 7/11 · 35%–85% | #5 · 6/12 · 25%–75% | 09-02→09-13 no significant change (z=-0.31); 09-13→09-20 no significant change (z=0.00) |
Sample 2
What models say: an AI glasses demo
12 models, Chinese and English, 1,656 answers in all: 576 describe a use case without naming any brand, and 720 ask about one of five brands by name. (Another 360 answers to a third named-brand question are left off this page; the end of this section explains why.) Every answer is coded sentence by sentence against a frozen list of aspects. What follows describes what the models said. It is not a verdict on the products, and not a measure of sales or market share.
Collected 2026-09-26 to 2026-09-27 · model-only panel of 12 models · codebook v1 · ebffb8ae97c3
Which brands come up in a use case
The question describes a use case and names no brand — for example, "I want a pair of AR glasses I can use as a portable big screen for movies. Any recommendations?" Each row is a brand. The two dots are the share of answers naming it when asked in Chinese and when asked in English; the thin lines are 95% intervals. The two numbers after a brand name count how often it was the first pick (Chinese · English).
- Asked in Chinese
- Asked in English
- 95% interval
Most often the first pick: XREAL, 42 times in Chinese and 41 in English (72 answers per language). Widest gap between the languages: RayNeo, named in 92% of answers in Chinese and 60% in English — 32 percentage points apart.
Prompts (Chinese): 「想买一副能当随身大屏看电影的 AR 眼镜，有什么推荐吗？」「坐飞机、躺着追剧用的 AR 眼镜，哪款好？」; (English): "I want a pair of AR glasses I can use as a portable big screen for movies. Any recommendations?" "What are good AR glasses for watching shows on a plane or in bed?"
Table view
| Brand | Asked in Chinese · Named (k/n) | 95% interval | First pick | Asked in English · Named (k/n) | 95% interval | First pick |
|---|---|---|---|---|---|---|
| XREAL | 72/72 · 100% | 95%–100% | 42 | 72/72 · 100% | 95%–100% | 41 |
| Rokid | 69/72 · 96% | 89%–99% | 0 | 64/72 · 89% | 80%–94% | 0 |
| RayNeo | 66/72 · 92% | 83%–96% | 4 | 43/72 · 60% | 48%–70% | 0 |
| VITURE | 43/72 · 60% | 48%–70% | 0 | 65/72 · 90% | 81%–95% | 6 |
| Apple | 13/72 · 18% | 11%–28% | — | 29/72 · 40% | 30%–52% | 0 |
| Meta (headsets) | 12/72 · 17% | 10%–27% | — | 26/72 · 36% | 26%–48% | 0 |
| Huawei | 10/72 · 14% | 8%–24% | — | 0/72 | not named | — |
- Asked in Chinese
- Asked in English
- 95% interval
Most often the first pick: Ray-Ban Meta, 3 times in Chinese and 23 in English (72 answers per language). Widest gap between the languages: INMO, named in 75% of answers in Chinese and 18% in English — 57 percentage points apart.
Prompts (Chinese): 「出国旅游想带一副能实时翻译的 AI 眼镜，有哪些可以选？」「带同声翻译功能的智能眼镜，哪些比较好用？」; (English): "I'm traveling abroad and want AI glasses that can translate conversations in real time. What are my options?" "Which smart glasses with live translation actually work well?"
Table view
| Brand | Asked in Chinese · Named (k/n) | 95% interval | First pick | Asked in English · Named (k/n) | 95% interval | First pick |
|---|---|---|---|---|---|---|
| Ray-Ban Meta | 67/72 · 93% | 85%–97% | 3 | 71/72 · 99% | 93%–100% | 23 |
| Even Realities | 46/72 · 64% | 52%–74% | 2 | 56/72 · 78% | 67%–86% | 3 |
| INMO | 54/72 · 75% | 64%–84% | 1 | 13/72 · 18% | 11%–28% | — |
| Rokid | 54/72 · 75% | 64%–84% | 1 | 30/72 · 42% | 31%–53% | 0 |
| RayNeo | 51/72 · 71% | 60%–80% | 1 | 33/72 · 46% | 35%–57% | 0 |
| XREAL | 33/72 · 46% | 35%–57% | 1 | 36/72 · 50% | 39%–61% | 2 |
| Solos | 14/72 · 19% | 12%–30% | — | 33/72 · 46% | 35%–57% | 1 |
- Asked in Chinese
- Asked in English
- 95% interval
Most often the first pick: Ray-Ban Meta, 48 times in Chinese and 64 in English (72 answers per language). Widest gap between the languages: RayNeo, named in 60% of answers in Chinese and 10% in English — 50 percentage points apart.
Prompts (Chinese): 「想要一副能随手拍照录像、记录日常的 AI 眼镜，买哪款好？」「带摄像头的智能眼镜，哪款比较值得买？」; (English): "I'd like AI glasses that can take photos and videos hands-free to capture everyday moments. Which should I buy?" "Which camera glasses are worth buying?"
Table view
| Brand | Asked in Chinese · Named (k/n) | 95% interval | First pick | Asked in English · Named (k/n) | 95% interval | First pick |
|---|---|---|---|---|---|---|
| Ray-Ban Meta | 72/72 · 100% | 95%–100% | 48 | 72/72 · 100% | 95%–100% | 64 |
| RayNeo | 43/72 · 60% | 48%–70% | 1 | 7/72 · 10% | 5%–19% | — |
| Rokid | 41/72 · 57% | 45%–68% | 0 | 14/72 · 19% | 12%–30% | — |
| Xiaomi | 40/72 · 56% | 44%–67% | 0 | 5/72 · 7% | 3%–15% | — |
| Snap | 3/72 · 4% | 1%–12% | — | 36/72 · 50% | 39%–61% | 0 |
| Huawei | 27/72 · 38% | 27%–49% | 0 | 2/72 · 3% | 1%–10% | — |
| Sharge | 23/72 · 32% | 22%–43% | 0 | 2/72 · 3% | 1%–10% | — |
- Asked in Chinese
- Asked in English
- 95% interval
Most often the first pick: Ray-Ban Meta, 11 times in Chinese and 57 in English (72 answers per language). Widest gap between the languages: Huawei, named in 86% of answers in Chinese and 6% in English — 81 percentage points apart.
Prompts (Chinese): 「我近视，想买一副平时能一直戴着的 AI 眼镜，有什么推荐？」「能配近视度数、适合日常佩戴的智能眼镜有哪些？」; (English): "I'm nearsighted and want AI glasses I can wear all day. What would you recommend?" "Which smart glasses work with prescription lenses for everyday wear?"
Table view
| Brand | Asked in Chinese · Named (k/n) | 95% interval | First pick | Asked in English · Named (k/n) | 95% interval | First pick |
|---|---|---|---|---|---|---|
| Ray-Ban Meta | 69/72 · 96% | 89%–99% | 11 | 72/72 · 100% | 95%–100% | 57 |
| Huawei | 62/72 · 86% | 76%–92% | 11 | 4/72 · 6% | 2%–13% | — |
| Even Realities | 26/72 · 36% | 26%–48% | 0 | 52/72 · 72% | 61%–81% | 0 |
| Rokid | 51/72 · 71% | 60%–80% | 0 | 20/72 · 28% | 19%–39% | 0 |
| RayNeo | 49/72 · 68% | 57%–78% | 3 | 9/72 · 13% | 7%–22% | — |
| Xiaomi | 46/72 · 64% | 52%–74% | 1 | 2/72 · 3% | 1%–10% | — |
| XREAL | 41/72 · 57% | 45%–68% | 0 | 39/72 · 54% | 43%–65% | 0 |
Method: a model-only panel of 12 models (panel snapshot a64ed9bbb94a), asked through developer APIs with no web access. Four use cases, two phrasings each, each phrasing asked three times. The Chinese and English phrasings were written separately rather than translated, and are computed separately: 72 answers per use case per language. No prompt names a brand, sets a format or asks for a fixed number of picks. Collected 2026-09-26 to 2026-09-27. "Named" is the share of answers naming the brand, with 95% Wilson intervals; a side where the brand was never named gets no dot. "First pick" counts answers recommending it as the first choice — at most one per answer, judged by coders under the codebook rules (double-coding agreement 97.5%). It is not reported, and shows as "—", on a side with fewer than 20 mentions. Each use case shows the seven brands with the highest rate in either language; sub-brands are merged into their parent brand.
What models say when asked by name
The brand is named in the question, in two phrasings: "Tell me about {brand}'s smart glasses." and "What are {brand} glasses like?" "Mentioned" is the share of answers that discuss an aspect; "with a drawback" is the share of those answers with at least one negative statement about it — the darker the cell, the higher the share. Each brand lists only the aspects mentioned in at least 50% of answers in at least one language.
Asked by name, the models bring up most aspects in more than half of their answers. The differences between brands, and between languages, show up more in the "with a drawback" column.
- Asked in Chinese
- Asked in English
| Aspect | Asked in Chinese | Asked in English | Language gap | ||
|---|---|---|---|---|---|
| Mentioned | With a drawback | Mentioned | With a drawback | ||
| Display & optics | 100% | 78% | 100% | 79% | |
| Comfort, weight & portability | 100% | 51% | 100% | 35% | |
| Connectivity, compatibility & software ecosystem | 100% | 89% | 100% | 81% | |
| Spatial tracking & controls | 99% | 63% | 100% | 65% | |
| Viewing & entertainment experience | 100% | 25% | 97% | 21% | |
| Productivity & work | 94% | 12% | 99% | 21% | |
| Brand & company | 99% | 4% | 90% | 0% | more in Chinese |
| Price & value | 81% | 33% | 72% | 14% | |
| Look & social acceptability | 63% | 2% | 78% | 36% | more in English |
| Prescription & lens options | 61% | 80% | 71% | 63% | |
| Battery & power | 43% | 87% | 56% | 48% | |
| Audio & calls | 31% | 36% | 61% | 41% | more in English |
Highest drawback share: Connectivity, compatibility & software ecosystem, at 89% in Chinese and 81% in English.
Table view
| Aspect | Asked in Chinese · Mentioned | 95% interval | With a drawback | Asked in English · Mentioned | 95% interval | With a drawback | Language gap |
|---|---|---|---|---|---|---|---|
| Display & optics | 72/72 · 100% | 95%–100% | 78% | 72/72 · 100% | 95%–100% | 79% | — |
| Comfort, weight & portability | 72/72 · 100% | 95%–100% | 51% | 72/72 · 100% | 95%–100% | 35% | — |
| Connectivity, compatibility & software ecosystem | 72/72 · 100% | 95%–100% | 89% | 72/72 · 100% | 95%–100% | 81% | — |
| Spatial tracking & controls | 71/72 · 99% | 93%–100% | 63% | 72/72 · 100% | 95%–100% | 65% | — |
| Viewing & entertainment experience | 72/72 · 100% | 95%–100% | 25% | 70/72 · 97% | 90%–99% | 21% | — |
| Productivity & work | 68/72 · 94% | 87%–98% | 12% | 71/72 · 99% | 93%–100% | 21% | — |
| Brand & company | 71/72 · 99% | 93%–100% | 4% | 65/72 · 90% | 81%–95% | 0% | more in Chinese |
| Price & value | 58/72 · 81% | 70%–88% | 33% | 52/72 · 72% | 61%–81% | 14% | — |
| Look & social acceptability | 45/72 · 63% | 51%–73% | 2% | 56/72 · 78% | 67%–86% | 36% | more in English |
| Prescription & lens options | 44/72 · 61% | 50%–72% | 80% | 51/72 · 71% | 60%–80% | 63% | — |
| Battery & power | 31/72 · 43% | 32%–55% | 87% | 40/72 · 56% | 44%–67% | 48% | — |
| Audio & calls | 22/72 · 31% | 21%–42% | 36% | 44/72 · 61% | 50%–72% | 41% | more in English |
- Asked in Chinese
- Asked in English
| Aspect | Asked in Chinese | Asked in English | Language gap | ||
|---|---|---|---|---|---|
| Mentioned | With a drawback | Mentioned | With a drawback | ||
| Display & optics | 100% | 65% | 99% | 54% | |
| Comfort, weight & portability | 100% | 44% | 99% | 30% | |
| Connectivity, compatibility & software ecosystem | 100% | 69% | 99% | 68% | |
| Viewing & entertainment experience | 97% | 19% | 97% | 14% | |
| Productivity & work | 85% | 18% | 97% | 14% | more in English |
| Brand & company | 100% | 8% | 82% | 0% | more in Chinese |
| Spatial tracking & controls | 81% | 62% | 86% | 66% | |
| Prescription & lens options | 81% | 22% | 78% | 16% | |
| AI assistant & smart features | 79% | 9% | 69% | 4% | |
| Look & social acceptability | 67% | 6% | 76% | 18% | |
| Audio & calls | 50% | 19% | 75% | 37% | more in English |
| Camera & capture | 61% | 7% | 63% | 4% | |
| Price & value | 71% | 28% | 53% | 11% | more in Chinese |
| Translation & captions | 63% | 0% | 53% | 0% | |
| Battery & power | 57% | 78% | 54% | 51% | |
Highest drawback share: Connectivity, compatibility & software ecosystem, at 69% in Chinese and 68% in English.
Table view
| Aspect | Asked in Chinese · Mentioned | 95% interval | With a drawback | Asked in English · Mentioned | 95% interval | With a drawback | Language gap |
|---|---|---|---|---|---|---|---|
| Display & optics | 72/72 · 100% | 95%–100% | 65% | 71/72 · 99% | 93%–100% | 54% | — |
| Comfort, weight & portability | 72/72 · 100% | 95%–100% | 44% | 71/72 · 99% | 93%–100% | 30% | — |
| Connectivity, compatibility & software ecosystem | 72/72 · 100% | 95%–100% | 69% | 71/72 · 99% | 93%–100% | 68% | — |
| Viewing & entertainment experience | 70/72 · 97% | 90%–99% | 19% | 70/72 · 97% | 90%–99% | 14% | — |
| Productivity & work | 61/72 · 85% | 75%–91% | 18% | 70/72 · 97% | 90%–99% | 14% | more in English |
| Brand & company | 72/72 · 100% | 95%–100% | 8% | 59/72 · 82% | 72%–89% | 0% | more in Chinese |
| Spatial tracking & controls | 58/72 · 81% | 70%–88% | 62% | 62/72 · 86% | 76%–92% | 66% | — |
| Prescription & lens options | 58/72 · 81% | 70%–88% | 22% | 56/72 · 78% | 67%–86% | 16% | — |
| AI assistant & smart features | 57/72 · 79% | 68%–87% | 9% | 50/72 · 69% | 58%–79% | 4% | — |
| Look & social acceptability | 48/72 · 67% | 55%–77% | 6% | 55/72 · 76% | 65%–85% | 18% | — |
| Audio & calls | 36/72 · 50% | 39%–61% | 19% | 54/72 · 75% | 64%–84% | 37% | more in English |
| Camera & capture | 44/72 · 61% | 50%–72% | 7% | 45/72 · 63% | 51%–73% | 4% | — |
| Price & value | 51/72 · 71% | 60%–80% | 28% | 38/72 · 53% | 41%–64% | 11% | more in Chinese |
| Translation & captions | 45/72 · 63% | 51%–73% | 0% | 38/72 · 53% | 41%–64% | 0% | — |
| Battery & power | 41/72 · 57% | 45%–68% | 78% | 39/72 · 54% | 43%–65% | 51% | — |
- Asked in Chinese
- Asked in English
| Aspect | Asked in Chinese | Asked in English | Language gap | ||
|---|---|---|---|---|---|
| Mentioned | With a drawback | Mentioned | With a drawback | ||
| Camera & capture | 94% | 28% | 93% | 9% | |
| Display & optics | 88% | 38% | 99% | 30% | more in English |
| Comfort, weight & portability | 97% | 43% | 89% | 33% | more in Chinese |
| AI assistant & smart features | 88% | 13% | 92% | 8% | |
| Audio & calls | 85% | 46% | 90% | 26% | |
| Brand & company | 83% | 20% | 90% | 15% | |
| Look & social acceptability | 78% | 23% | 94% | 25% | more in English |
| Connectivity, compatibility & software ecosystem | 81% | 48% | 90% | 40% | |
| Availability & regional access | 69% | 98% | 96% | 99% | more in English |
| Translation & captions | 78% | 2% | 86% | 5% | |
| Battery & power | 75% | 65% | 63% | 56% | |
| Price & value | 75% | 24% | 60% | 19% | |
| Prescription & lens options | 68% | 16% | 57% | 15% | |
| Spatial tracking & controls | 42% | 10% | 79% | 12% | more in English |
Highest drawback share: Availability & regional access, at 98% in Chinese and 99% in English.
Table view
| Aspect | Asked in Chinese · Mentioned | 95% interval | With a drawback | Asked in English · Mentioned | 95% interval | With a drawback | Language gap |
|---|---|---|---|---|---|---|---|
| Camera & capture | 68/72 · 94% | 87%–98% | 28% | 67/72 · 93% | 85%–97% | 9% | — |
| Display & optics | 63/72 · 88% | 78%–93% | 38% | 71/72 · 99% | 93%–100% | 30% | more in English |
| Comfort, weight & portability | 70/72 · 97% | 90%–99% | 43% | 64/72 · 89% | 80%–94% | 33% | more in Chinese |
| AI assistant & smart features | 63/72 · 88% | 78%–93% | 13% | 66/72 · 92% | 83%–96% | 8% | — |
| Audio & calls | 61/72 · 85% | 75%–91% | 46% | 65/72 · 90% | 81%–95% | 26% | — |
| Brand & company | 60/72 · 83% | 73%–90% | 20% | 65/72 · 90% | 81%–95% | 15% | — |
| Look & social acceptability | 56/72 · 78% | 67%–86% | 23% | 68/72 · 94% | 87%–98% | 25% | more in English |
| Connectivity, compatibility & software ecosystem | 58/72 · 81% | 70%–88% | 48% | 65/72 · 90% | 81%–95% | 40% | — |
| Availability & regional access | 50/72 · 69% | 58%–79% | 98% | 69/72 · 96% | 89%–99% | 99% | more in English |
| Translation & captions | 56/72 · 78% | 67%–86% | 2% | 62/72 · 86% | 76%–92% | 5% | — |
| Battery & power | 54/72 · 75% | 64%–84% | 65% | 45/72 · 63% | 51%–73% | 56% | — |
| Price & value | 54/72 · 75% | 64%–84% | 24% | 43/72 · 60% | 48%–70% | 19% | — |
| Prescription & lens options | 49/72 · 68% | 57%–78% | 16% | 41/72 · 57% | 45%–68% | 15% | — |
| Spatial tracking & controls | 30/72 · 42% | 31%–53% | 10% | 57/72 · 79% | 68%–87% | 12% | more in English |
- Asked in Chinese
- Asked in English
| Aspect | Asked in Chinese | Asked in English | Language gap | ||
|---|---|---|---|---|---|
| Mentioned | With a drawback | Mentioned | With a drawback | ||
| Brand & company | 94% | 7% | 93% | 3% | |
| AI assistant & smart features | 86% | 14% | 92% | 6% | |
| Translation & captions | 68% | 0% | 81% | 2% | |
| Connectivity, compatibility & software ecosystem | 68% | 31% | 72% | 21% | |
| Availability & regional access | 54% | 100% | 69% | 90% | |
| Camera & capture | 44% | 3% | 71% | 4% | more in English |
| Audio & calls | 36% | 15% | 67% | 4% | more in English |
| Display & optics | 35% | 36% | 53% | 18% | more in English |
Highest drawback share: Availability & regional access, at 100% in Chinese and 90% in English.
9 of the answers above describe "Alibaba Qwen" as the Qwen language model and say nothing about glasses (5 in Chinese, 4 in English); they stay in the denominator. Of the 576 use-case answers above, 2 in Chinese and 0 in English name Alibaba Qwen.
How to read thisThese glasses launched recently, and the most likely reason is that these models' training data predates the launch. What this measures is that the product is not yet in the models' memory; it cannot be read as the models rating it poorly.
Table view
| Aspect | Asked in Chinese · Mentioned | 95% interval | With a drawback | Asked in English · Mentioned | 95% interval | With a drawback | Language gap |
|---|---|---|---|---|---|---|---|
| Brand & company | 68/72 · 94% | 87%–98% | 7% | 67/72 · 93% | 85%–97% | 3% | — |
| AI assistant & smart features | 62/72 · 86% | 76%–92% | 14% | 66/72 · 92% | 83%–96% | 6% | — |
| Translation & captions | 49/72 · 68% | 57%–78% | 0% | 58/72 · 81% | 70%–88% | 2% | — |
| Connectivity, compatibility & software ecosystem | 49/72 · 68% | 57%–78% | 31% | 52/72 · 72% | 61%–81% | 21% | — |
| Availability & regional access | 39/72 · 54% | 43%–65% | 100% | 50/72 · 69% | 58%–79% | 90% | — |
| Camera & capture | 32/72 · 44% | 34%–56% | 3% | 51/72 · 71% | 60%–80% | 4% | more in English |
| Audio & calls | 26/72 · 36% | 26%–48% | 15% | 48/72 · 67% | 55%–77% | 4% | more in English |
| Display & optics | 25/72 · 35% | 25%–46% | 36% | 38/72 · 53% | 41%–64% | 18% | more in English |
- Asked in Chinese
- Asked in English
| Aspect | Asked in Chinese | Asked in English | Language gap | ||
|---|---|---|---|---|---|
| Mentioned | With a drawback | Mentioned | With a drawback | ||
| Camera & capture | 100% | 49% | 100% | 51% | |
| Audio & calls | 100% | 29% | 100% | 44% | |
| Look & social acceptability | 100% | 6% | 100% | 10% | |
| Battery & power | 100% | 75% | 100% | 67% | |
| AI assistant & smart features | 99% | 34% | 100% | 14% | |
| Brand & company | 100% | 4% | 97% | 6% | |
| Privacy & safety concerns | 93% | 79% | 96% | 84% | |
| Connectivity, compatibility & software ecosystem | 89% | 64% | 88% | 60% | |
| Prescription & lens options | 82% | 27% | 92% | 23% | |
| Translation & captions | 83% | 7% | 86% | 5% | |
| Price & value | 85% | 49% | 74% | 26% | |
| Display & optics | 74% | 83% | 75% | 76% | |
| Availability & regional access | 97% | 100% | 51% | 89% | more in Chinese |
| Comfort, weight & portability | 81% | 24% | 60% | 28% | more in Chinese |
| Build, reliability & support | 47% | 50% | 50% | 50% | |
Highest drawback share: Availability & regional access, at 100% in Chinese and 89% in English.
Table view
| Aspect | Asked in Chinese · Mentioned | 95% interval | With a drawback | Asked in English · Mentioned | 95% interval | With a drawback | Language gap |
|---|---|---|---|---|---|---|---|
| Camera & capture | 72/72 · 100% | 95%–100% | 49% | 72/72 · 100% | 95%–100% | 51% | — |
| Audio & calls | 72/72 · 100% | 95%–100% | 29% | 72/72 · 100% | 95%–100% | 44% | — |
| Look & social acceptability | 72/72 · 100% | 95%–100% | 6% | 72/72 · 100% | 95%–100% | 10% | — |
| Battery & power | 72/72 · 100% | 95%–100% | 75% | 72/72 · 100% | 95%–100% | 67% | — |
| AI assistant & smart features | 71/72 · 99% | 93%–100% | 34% | 72/72 · 100% | 95%–100% | 14% | — |
| Brand & company | 72/72 · 100% | 95%–100% | 4% | 70/72 · 97% | 90%–99% | 6% | — |
| Privacy & safety concerns | 67/72 · 93% | 85%–97% | 79% | 69/72 · 96% | 89%–99% | 84% | — |
| Connectivity, compatibility & software ecosystem | 64/72 · 89% | 80%–94% | 64% | 63/72 · 88% | 78%–93% | 60% | — |
| Prescription & lens options | 59/72 · 82% | 72%–89% | 27% | 66/72 · 92% | 83%–96% | 23% | — |
| Translation & captions | 60/72 · 83% | 73%–90% | 7% | 62/72 · 86% | 76%–92% | 5% | — |
| Price & value | 61/72 · 85% | 75%–91% | 49% | 53/72 · 74% | 62%–82% | 26% | — |
| Display & optics | 53/72 · 74% | 62%–82% | 83% | 54/72 · 75% | 64%–84% | 76% | — |
| Availability & regional access | 70/72 · 97% | 90%–99% | 100% | 37/72 · 51% | 40%–63% | 89% | more in Chinese |
| Comfort, weight & portability | 58/72 · 81% | 70%–88% | 24% | 43/72 · 60% | 48%–70% | 28% | more in Chinese |
| Build, reliability & support | 34/72 · 47% | 36%–59% | 50% | 36/72 · 50% | 39%–61% | 50% | — |
Method: two named-brand phrasings, written separately in each language — Chinese 「介绍一下{brand}的智能眼镜。」「{brand}的眼镜怎么样？」, English "Tell me about {brand}'s smart glasses." "What are {brand} glasses like?". {brand} is the brand's Chinese name in the Chinese prompts and its English name in the English ones. Per brand and language: 2 phrasings × 3 repeats × 12 models = 72 answers; panel and collection dates as above. Every answer is coded sentence by sentence against the frozen codebook v1: 18 aspects plus "other", SHA-256 prefix ebffb8ae97c3. "Mentioned" is the share of answers with at least one sentence on that aspect, with 95% Wilson intervals in the table view; "with a drawback" is the share of those answers with at least one negative statement, not reported below 20 mentions. "Language gap" is a two-proportion test, run only when both sides have at least 20 mentions and shown only at p < 0.05. Reliability: 336 randomly drawn answers were coded again independently by a second coder; Cohen's κ across the 19 codes ranges from 0.87 to 0.98, and agreement on negative statements is 97.2%.
More samples
Mention-rate change
data being collectedTwo deeply sampled periods on the same frozen questions, with a three-way verdict for every cell. It will appear here once both periods are in.
The compiled data on this page (ranks, model coverage, mention rates, intervals, change markers) is released under CC BY 4.0; please credit ask10ais and the measurement date. Rights in the raw model answers belong to each model provider; we keep them unaltered so they can be checked.
This record measures what large language models answered to one question at one moment. It is not an assessment of product quality, market share or corporate standing. Model output may contain errors or fabrications; this record preserves the raw responses as returned and does not endorse their content. It is not investment or procurement advice.