profile-img
lmsys.org

@lmsysorg

Large Model Systems Organization. We created Vicuna and Chatbot Arena! Compare 30+ LLMs (GPT-4/Claude/Llamas) side-by-side at https://t.co/IDFeIDIOtm

calendar_today30-03-2023 09:56:38

460 Tweets

43,2K Followers

171 Following

lmsys.org(@lmsysorg) 's Twitter Profile Photo
lmsys.org
@lmsysorg

2 weeks ago

Breaking news — gpt2-chatbots result is now out!

gpt2-chatbots have just surged to the top, surpassing all the models by a significant gap (~50 Elo). It has become the strongest model ever in the Arena!

With improvement across all boards, especially reasoning & coding

Breaking news — gpt2-chatbots result is now out! gpt2-chatbots have just surged to the top, surpassing all the models by a significant gap (~50 Elo). It has become the strongest model ever in the Arena! With improvement across all boards, especially reasoning & coding
account_circle
lmsys.org(@lmsysorg) 's Twitter Profile Photo
lmsys.org
@lmsysorg

2 weeks ago

Confidence intervals chart — huge gap against previous top-5 models.

Confidence intervals chart — huge gap against previous top-5 models.
account_circle
lmsys.org(@lmsysorg) 's Twitter Profile Photo
lmsys.org
@lmsysorg

2 weeks ago

In more challenging Coding Arena, we see even bigger gap (~100 Elo)!

In more challenging Coding Arena, we see even bigger gap (~100 Elo)!
account_circle
lmsys.org(@lmsysorg) 's Twitter Profile Photo
lmsys.org
@lmsysorg

2 weeks ago

Significantly higher win-rate against all other models.
e.g., ~80% win-rate vs GPT-4 (June) in non-tie battles.

Significantly higher win-rate against all other models. e.g., ~80% win-rate vs GPT-4 (June) in non-tie battles.
account_circle