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
Every model the catalog carries a published Joke Arena value for, ranked by that value.
Kaggle's Joke Arena asks models for an original joke on 100 topics, and people vote for the funnier one. 18 models in the ModelScale catalog have a published value.
Each point is a model with a published Joke Arena value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
18 models with a published Joke Arena 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 | Relative rating fitted to human pairwise votes |
|---|---|---|---|
| 1 | OpenAI | 1568 | |
| 2 | Anthropic | 1548 | |
| 3 | 1540 |
What the benchmark measures and how its value is published, from the source’s own record.
Joke Arena is an external benchmark published by BenchLM. Kaggle's Joke Arena asks models for an original joke on 100 topics, and people vote for the funnier one. Its evaluation set is described by the source as 100 joke prompts in 10 categories, rated by human votes. The source rates its difficulty as humour writing.
Each model's result is published by BenchLM as relative rating fitted to human pairwise votes. ModelScale prints that value to 0 decimal places, exactly as published, and never converts or rescales it. This page ranks the 18 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 | Z.AI | 1539 |
| 5 | Anthropic | 1536 |
| 6 | OpenAI | 1520 |
| 7 | Anthropic | 1505 |
| 8 | OpenAI | 1502 |
| 8 | OpenAI | 1502 |
| 10 | 1494 |
| 11 | xAI | 1482 |
| 12 | OpenAI | 1481 |
| 13 | OpenAI | 1478 |
| 14 | xAI | 1464 |
| 15 | Alibaba | 1461 |
| 16 | Anthropic | 1459 |
| 16 | Anthropic | 1459 |
| 18 | Anthropic | 1458 |
Evidence key: ObservedLast good
Rows are ordered by the value BenchLM 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.