Benchmark checker Multilingual

Is EuroEval still meaningful?

Still useful Checked against the report, 5 November 2026

Yes, for European language skill. It does not test legal conduct.

Why still useful: It is the live, maintained European benchmark; version 18.3.0 shipped on 8 October 2026.^{[1][1] Site summary EuroEval Primary}^{[2][2] EuroEval repository Primary}

What “Still useful” means

Known problems

Its "European Values" tasks are the nearest European behavioural signal, but they are not legal tests.^{[1][1] Site summary EuroEval Primary}
Its top rows show no Anthropic model, probably because of which models it ran.^{[3][3] Dutch generative leaderboard EuroEval Primary}

What to look at instead

The report names no direct successor.

Where it appears in the report

Chapter 1, Benchmarks

Domain benchmarks test expert skill, not lawful conduct
Figure 11. Few benchmarks test conduct under a named law
Table 15. Multilingual and European benchmarks, November 2026 in “Go deeper: Multilingual and European benchmarks (7)”

Sources

Copied from the chapter source lists. Each number links back to the source in its chapter.

[1]
Site summary, EuroEval. It lists the languages, boards and tasks. Primary euroeval.com Chapter 1, source 160
[2]
EuroEval repository. It names the maintainer, licence, sponsors and releases. Primary github.com Chapter 1, source 161
[3]
Dutch generative leaderboard, EuroEval. It ranks models on Dutch tasks. Primary euroeval.com Chapter 1, source 162
[4]
German generative leaderboard, EuroEval. It ranks models on German tasks. Primary euroeval.com Chapter 1, source 163