Robot vacuums
Which brand 5 AI models pick, with intervals.
Fast panel · Panel, equal weight
- anthropic/claude-haiku-5.5 @ claude-haiku-5.5 via OpenRouter (20%)
- openai/gpt-6-luna @ gpt-6-luna via OpenRouter (20%)
- google/gemini-4-argon @ gemini-4-argon via OpenRouter (20%)
- mistralai/mistral-large-4-0 @ mistral-large-4-0 via OpenRouter (20%)
- deepseek/deepseek-v4.1-flash @ deepseek-v4.1-flash via OpenRouter (20%)
ClearRoborock stands alone in the top tier for these 5 API models.
Tier 1
- Roborock79%95% interval 66–90%plausible rank 1
Tier 2
- Dreame39%95% interval 29%–< 50%plausible rank 2–7
- iRobot32%95% interval 20–45%plausible rank 2–8
- eufy29%95% interval 16–43%plausible rank 2–8
- Narwal22%95% interval 13–31%plausible rank 2–8
- Shark16%95% interval 8–26%plausible rank 2–10
- ECOVACS16%95% interval 8–25%plausible rank 3–10
- Tapo12%95% interval 4–25%plausible rank 2–10
Tier 3
- SwitchBot4%95% interval 1–12%plausible rank 6–10
- Samsung3%95% interval 0–9%plausible rank 6–10
Drag across the chart to probe a share.
Changes since the previous edition
The judge panel changed since the previous edition, so shares are not compared between the two.
- google/gemini-4-argon @ gemini-4-argon via OpenRouter: added in this edition
- mistralai/mistral-large-4-0 @ mistral-large-4-0 via OpenRouter: same model, new judge settings
This note was added after publication from the published editions; the signed snapshot is unchanged. Corrections log
ClearRoborock stands alone in the top tier for these 5 API models.
| Brand | Choice share | 95% interval | Plausible rank | Tier |
|---|---|---|---|---|
| Roborock | 79% | 66–90% | 1 | Tier 1 |
| Dreame | 39% | 29%–< 50% | 2–7 | Tier 2 |
| iRobot | 32% | 20–45% | 2–8 | Tier 2 |
| eufy | 29% | 16–43% | 2–8 | Tier 2 |
| Narwal | 22% | 13–31% | 2–8 | Tier 2 |
| Shark | 16% | 8–26% | 2–10 | Tier 2 |
| ECOVACS | 16% | 8–25% | 3–10 | Tier 2 |
| Tapo | 12% | 4–25% | 2–10 | Tier 2 |
| SwitchBot | 4% | 1–12% | 6–10 | Tier 3 |
| Samsung | 3% | 0–9% | 6–10 | Tier 3 |
How to read this
- Choice share: how often models pick a brand when it appears in a shortlist of 4. Chance is 25%, the dashed line.
- The band is the 95% interval; the bright tick is the estimate. Overlapping bands are not a strict order.
- Tiers are published: a tier groups brands not distinguishable from that tier's leader. Same tier does not mean equal.
- Rank shows the plausible rank range, not a strict place.
- Drag across the chart to probe a share and see which intervals contain it.
- Brand names only: the models saw each option as a brand name, with no product facts, and answered from their own knowledge.
- Fast panel: the fast or small current models of each vendor, not the full flagship panel.
- Each page shows the latest edition. Dates are in the downloadable JSON.
Panel and conditions
Each edition puts the brands of the category to a panel of AI models, called through their APIs, in randomized shortlists of 4 (chance is 25%). Each option was a brand name only, with no product facts, prices or any commercial data, so a brand's share measures which brand the models recommend from their own knowledge for the shopper needs of the scenarios. Shares come with 95% intervals; tiers and rank intervals are simultaneous at the same level.
- anthropic/claude-haiku-5.5 @ claude-haiku-5.5 via OpenRouter (20%)
- openai/gpt-6-luna @ gpt-6-luna via OpenRouter (20%)
- google/gemini-4-argon @ gemini-4-argon via OpenRouter (20%)
- mistralai/mistral-large-4-0 @ mistral-large-4-0 via OpenRouter (20%)
- deepseek/deepseek-v4.1-flash @ deepseek-v4.1-flash via OpenRouter (20%)
Conditions: closed-book API calls via OpenRouter (pinned providers); web search, grounding, tools and retrieval off; forced choice; 2 paraphrases per scenario in English.
Methodology: ACM-1.0-chart-draft (draft rules for movement, significance and events, awaiting review).