Skip to main content
ModelScale

Agentic benchmark

MEWC leaderboard

Multi-Environment Web Challenge. Every model the catalog carries a published MEWC value for, ranked by that value.

CategoryAgentic
MeasureBrowser task completion
TasksWeb-agent tasks
DifficultyOpen-web agent workflows

A benchmark that evaluates AI agents on multi-environment web challenges, testing navigation and task completion across diverse live web environments.

No model in the catalog has a published MEWC score.

The benchmark is defined by MiniMax M2.5 benchmark release surface, but the catalog carries no value for it yet. An absent value is shown as absent here rather than as a zero.

All leaderboards · Agentic capability leaderboard