Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
external benchmark
Cookie-Bench: Continuous On-screen Key Interaction Evaluation for Web Generation. Every model the catalog carries a published Cookie-Bench value for, ranked by that value.
A 1,000-query web development benchmark that judges generated front-end apps by deploying them and letting an evaluator agent use them, scoring functionality and aesthetics without reference implementations or test suites. 7 models in the ModelScale catalog have a published value.
Each point is a model with a published Cookie-Bench value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
7 models with a published Cookie-Bench value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Benchmark score (0-100) |
|---|---|---|---|
| 1 | Anthropic | 91.5 | |
| 2 | OpenAI | 89.7 | |
| 3 | Tencent | 88.8 |
What the benchmark measures and how its value is published, from the source’s own record.
Cookie-Bench (Cookie-Bench: Continuous On-screen Key Interaction Evaluation for Web Generation) is an external benchmark published by Cookie-Bench: Continuous On-screen Key Interaction Evaluation for Web Generation. A 1,000-query web development benchmark that judges generated front-end apps by deploying them and letting an evaluator agent use them, scoring functionality and aesthetics without reference implementations or test suites. Its evaluation set is described by the source as 1,000 web development queries (React setting). The source rates its difficulty as interactive web apps and static pages across three tiers.
Each model's result is published by Cookie-Bench: Continuous On-screen Key Interaction Evaluation for Web Generation as benchmark score (0-100). ModelScale prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 7 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
| 4 | Baidu | 88.2 |
| 5 | DeepSeek | 87.9 |
| 6 | Z.AI | 80.4 |
| 7 | Z.AI | 76.9 |
Evidence key: Observed
Rows are ordered by the value Cookie-Bench: Continuous On-screen Key Interaction Evaluation for Web Generation published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.