This model is a merged model based on Qwen/Qwen2.5-7B-Instruct using a novel model merging technique.
Performance (Self-Tested on A100)
The following results are obtained using batch_size=6 on an A100 GPU. Official results are pending submission to open_llm_leaderboard
.
IFEVAL | BBH | MATH | GPQA | MUSR | MMLU-PRO | AVG |
---|---|---|---|---|---|---|
76.4 | 36.09 | 48.04 | 8.05 | 14.24 | 37.7 | 36.75 |
Note: These results will be updated once officially verified.
Recipe Coming Soon
We will release details on the merging technique and methodology soon. Stay tuned! 🚀