Dicta
We track 3 benchmarks that Dicta made or helped make.
0 in use 3 with public code 2 with a leaderboard here
Benchmarks
| Name | Kind | Released | Best score | Status |
|---|---|---|---|---|
| Open Hebrew LLM Leaderboard Models answer questions, resolve pronouns, classify sentiment and inference, translate passages, and answer trivia in Hebrew. | Knowledge in many languages | Mar 2024 | | Not in cards |
| Hebrew LLM Chat Leaderboard Models summarize documents, translate into Hebrew, answer trivia, resolve pronouns, and restore vowel marks. The suite reports each task separately. | Chat and instructions in many languages | Nov 2025 | | Not in cards |
| Hebrew Diacritization Test Corpora Models restore vowel and consonant marks in Hebrew prose, rabbinic commentary and liturgical poetry. The reference also identifies spelling letters that vocalized text should omit. | Reading in many languages | Jan 2020 | Not in cards |