Prometheus is evaluated on the Azerbaijani questions of TUMLU, a benchmark written natively in eight Turkic languages, so its Azerbaijani is measured on real Azerbaijani material rather than on translated tests.
It is easy to claim a model speaks Azerbaijani; it is harder to prove it. Prometheus is evaluated on TUMLU, a benchmark of 38,139 questions written natively in eight Turkic languages across 11 academic subjects. Its Azerbaijani questions were written in Azerbaijani, not translated from English, so the test measures the language as people actually use it.
Evaluated on TUMLU’s native Azerbaijani questions.
TUMLU covers 11 academic subjects in eight Turkic languages, Azerbaijani among them.
No translated questions: each was written in its own language.
Provides an objective measure of Azerbaijani ability.
Backs a model built natively for Azerbaijani and run on your own infrastructure.
TUMLU is a benchmark of 38,139 questions written natively in eight Turkic languages, Azerbaijani among them, across 11 academic subjects. Prometheus is evaluated on its Azerbaijani questions.
No. TUMLU’s Azerbaijani questions were written in Azerbaijani, with no translated content.
It replaces marketing claims with an objective measure of how well the model actually handles native Azerbaijani across many disciplines.
The first large language model built natively for Azerbaijani — not adapted from English.
Runs fully inside your own network — data never leaves your infrastructure.
Three model sizes to match your accuracy needs and hardware budget.
Handles the ə character and agglutinative morphology — 4.6× more efficient on Azerbaijani text.
See the complete product: problem, features, how it works and deployment.
Request a demo to see Prometheus running on-premise — Azerbaijani understanding, model sizing and API integration, end to end.