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
Every model the catalog carries a published MLE-Bench Lite value for, ranked by that value.
A lightweight machine-learning competition benchmark that measures whether models can iteratively train, evaluate, and improve ML systems in low-resource settings. 4 models in the ModelScale catalog have a published value.
Each point is a model with a published MLE-Bench Lite value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
4 models with a published MLE-Bench Lite 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 | Autonomous iterative ML optimization |
|---|---|---|---|
| 1 | Shanghai Artificial Intelligence Laboratory | 86.2 | |
| 2 | MiniMax | 66.6 | |
| 3 | InternScience |
What the benchmark measures and how its value is published, from the source’s own record.
MLE-Bench Lite is an agentic benchmark published by MiniMax M2.7: Early Echoes of Self-Evolution. A lightweight machine-learning competition benchmark that measures whether models can iteratively train, evaluate, and improve ML systems in low-resource settings. Its evaluation set is described by the source as low-resource ML competitions. The source rates its difficulty as agentic machine learning.
Each model's result is published by MiniMax M2.7: Early Echoes of Self-Evolution as autonomous iterative ML optimization. ModelScale prints that value exactly as published, and never converts or rescales it. This page ranks the 4 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 | InternScience | 22.7 |
Evidence key: Observed
Rows are ordered by the value MiniMax M2.7: Early Echoes of Self-Evolution 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.