Spaces:

open-llm-leaderboard

/

open_llm_leaderboard

like 12.2k

Running on CPU Upgrade

I just tweeted something with those numbers from here yesterday! 😅 I hope they were accurate! https://x.com/MaziyarPanahi/status/1800168264565395534

Hm interesting, it would seem our dataset parser (here) removed them. We're having the same issue with llama 2 - 70B.
(You can see the current version of the dataset behind the leaderboard here).

I'll investigate a bit later, but if one of you wants to take a look before, please be my guest :)

No update yet? What is the use of this rating then if there isn't a comparison with latest base models?

You'll now find Qwen2 models on the leaderboard v2! Cheers! :)