paulml

/

DPOB-INMTOB-7B

like 2

Just wanted to express some support for this model. It still has the highest average rating on the leaderboards for its inference speed, which is 120 tokens/s on my RTX 3090. Keep it up!

Thanks a lot! What leaderboard are you talking about?

https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard