ceferisbarov/TUMLU
TUMLU: A Unified and Native Language Understanding Benchmark for Turkic Languages
This project helps evaluate how well different large language models understand Turkic languages. It takes a model and a specific Turkic language as input and outputs a score indicating the model's performance on middle- and high-school level comprehension questions. This is for researchers, linguists, or educators who want to compare and understand the capabilities of AI models for Turkic language applications.
No commits in the last 6 months.
Use this if you need to benchmark and compare the natural language understanding capabilities of various AI models across multiple Turkic languages using native, non-translated questions.
Not ideal if you're looking for a tool to build an application or directly improve a model's performance; it's designed for evaluation and comparison.
Stars
25
Forks
1
Language
Python
License
—
Category
Last pushed
Feb 25, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/ceferisbarov/TUMLU"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
hltcoe/turkle
Django-based clone of Amazon's Mechanical Turk service running in your local environment.
emres/turkish-deasciifier
Turkish deasciifier in Python based on Deniz Yüret's turkish-mode for Emacs
ahmetaa/zemberek-nlp
NLP tools for Turkish.
ooguz/turkce-kufur-karaliste
Türkçe için bir kara liste (blacklist)
brolin59/trnlp
TÜRKÇE İÇİN DOĞAL DİL İŞLEME ARAÇLARI