malhajar/OpenLLMTurkishLeaderboard_v0.2 · Extending the coverage of the leaderboard

The current leaderboard does a great job of comparing models on Turkish-specific benchmarks, but I think it could be even better with a few more strong models added.
There are several SoTA models that perform really well but aren’t currently listed. For example, these models have shown solid results on the Turkish Evaluation Benchmark:
https://huggingface.co/ytu-ce-cosmos/Turkish-Gemma-9b-v0.1#%F0%9F%93%8A-turkish-evaluation-benchmark-results-via-malhajar17lm-evaluation-harness_turkish
(They were already got tested by the research group using Turkish datasets of this space)
A few models worth adding:

google/gemma-3-12b-it
Qwen/Qwen2.5-14B-it

Adding these could make the leaderboard more complete and useful for the community. Thanks.