Soft drinks
Which brand 10 AI models pick, with intervals.
Panel, equal weight
- anthropic/claude-opus-5.5 @ claude-opus-5.5 via OpenRouter (10%)
- openai/gpt-6.1-sol @ gpt-6.1-sol via OpenRouter (10%)
- google/gemini-4-argon @ gemini-4-argon via OpenRouter (10%)
- deepseek/deepseek-v4-pro @ deepseek-v4-pro via OpenRouter (10%)
- mistralai/mistral-large-4-0 @ mistral-large-4-0 via OpenRouter (10%)
- meta-llama/llama-4-maverick-17b-128e-instruct @ llama-4-maverick-17b-128e-instruct via OpenRouter (10%)
- moonshotai/kimi-k3.1-pro @ kimi-k3.1-pro via OpenRouter (10%)
- z-ai/glm-5.3 @ glm-5.3 via OpenRouter (10%)
- qwen/qwen3.8-max-prime @ qwen3.8-max-prime via OpenRouter (10%)
- x-ai/grok-4.7 @ grok-4.7 via OpenRouter (10%)
Too closeNo single favourite: Coca-Cola, Dr Pepper, Mountain Dew, OLIPOP, Pepsi, and Sprite share the top tier for these 10 API models.
Tier 1
- Coca-Cola51%95% interval 38–63%plausible rank 1–6
- Pepsi39%95% interval 27–52%plausible rank 1–12
- Dr Pepper35%95% interval 25–46%plausible rank 1–12
- Sprite31%95% interval 20–42%plausible rank 1–12
- Mountain Dew26%95% interval 16–37%plausible rank 1–12
- OLIPOP22%95% interval 12–33%plausible rank 1–12
Tier 2
- A&W20%95% interval 12–30%plausible rank 2–12
- poppi17%95% interval 9–27%plausible rank 2–12
- 7UP15%95% interval 7–25%plausible rank 2–12
- Canada Dry15%95% interval 7–26%plausible rank 2–12
- Sunkist15%95% interval 7–25%plausible rank 2–12
- Fanta15%95% interval 7–25%plausible rank 2–12
Drag across the chart to probe a share.
Too closeNo single favourite: Coca-Cola, Dr Pepper, Mountain Dew, OLIPOP, Pepsi, and Sprite share the top tier for these 10 API models.
| Brand | Choice share | 95% interval | Plausible rank | Tier |
|---|---|---|---|---|
| Coca-Cola | 51% | 38–63% | 1–6 | Tier 1 |
| Pepsi | 39% | 27–52% | 1–12 | Tier 1 |
| Dr Pepper | 35% | 25–46% | 1–12 | Tier 1 |
| Sprite | 31% | 20–42% | 1–12 | Tier 1 |
| Mountain Dew | 26% | 16–37% | 1–12 | Tier 1 |
| OLIPOP | 22% | 12–33% | 1–12 | Tier 1 |
| A&W | 20% | 12–30% | 2–12 | Tier 2 |
| poppi | 17% | 9–27% | 2–12 | Tier 2 |
| 7UP | 15% | 7–25% | 2–12 | Tier 2 |
| Canada Dry | 15% | 7–26% | 2–12 | Tier 2 |
| Sunkist | 15% | 7–25% | 2–12 | Tier 2 |
| Fanta | 15% | 7–25% | 2–12 | Tier 2 |
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.
- Flagship panel: one current flagship model per major vendor, plus the most-used models.
- 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-opus-5.5 @ claude-opus-5.5 via OpenRouter (10%)
- openai/gpt-6.1-sol @ gpt-6.1-sol via OpenRouter (10%)
- google/gemini-4-argon @ gemini-4-argon via OpenRouter (10%)
- deepseek/deepseek-v4-pro @ deepseek-v4-pro via OpenRouter (10%)
- mistralai/mistral-large-4-0 @ mistral-large-4-0 via OpenRouter (10%)
- meta-llama/llama-4-maverick-17b-128e-instruct @ llama-4-maverick-17b-128e-instruct via OpenRouter (10%)
- moonshotai/kimi-k3.1-pro @ kimi-k3.1-pro via OpenRouter (10%)
- z-ai/glm-5.3 @ glm-5.3 via OpenRouter (10%)
- qwen/qwen3.8-max-prime @ qwen3.8-max-prime via OpenRouter (10%)
- x-ai/grok-4.7 @ grok-4.7 via OpenRouter (10%)
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).