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.
Agentic benchmark
Agents' Last Exam, Linux subset. Every model the catalog carries a published Agents' Last Exam (Linux) value for, ranked by that value.
The Linux subset of Agents' Last Exam, an end-to-end agent task benchmark, as reported by Meta. 3 models in the ModelScale catalog have a published value.
Each point is a model with a published Agents' Last Exam (Linux) value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
3 models with a published Agents' Last Exam (Linux) 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 | Provider-reported task score |
|---|---|---|---|
| 1 | Meta | 61.6 | |
| 2 | OpenAI | 57.1 | |
| 3 | Anthropic | 57.0 |
What the benchmark measures and how its value is published, from the source’s own record.
Agents' Last Exam (Linux) (Agents' Last Exam, Linux subset) is an agentic benchmark published by Introducing Muse Ember. The Linux subset of Agents' Last Exam, an end-to-end agent task benchmark, as reported by Meta. Its evaluation set is described by the source as linux agent tasks (subset size not published). The source rates its difficulty as advanced agentic work.
Each model's result is published by Introducing Muse Ember as provider-reported task score. ModelScale prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 3 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.
Evidence key: Observed
Rows are ordered by the value Introducing Muse Ember 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.