Playing Chess will never be the same again. Meet ChessArena - The only open-source Chess Benchmark for LLMs

nice work

thank you!!

Try yourself -

Looks Amazing !!

Thanks, do try yourself.

This is interesting! Watching Claude beat OpenAI here -

@AnthropicAI Claude 3.5 Sonnet is the Best? Who knows, we have to check leaderboard first!

Looks good 👍

Glad you liked it!

@ai_alok 😢😢😢😢😢😢

Looks Cool

Watching LLMs playing chess is FUN man!

🔥

Checkmate.

Really impressive 🥳

It's crazy how this models makes silly mistakes.

Strategy, cognitive, planning, problem solving , memory and prioritization. All being tested in one game . Nice one Rohit

You summed up everything.

@aragunde

Thanks for sharing 🔥

You'll love it more than anyone!

@GothamChess content?

@GothamChess

Amazing 👌

Thanks

wdyt?

.@theo check this out.

@theo @theo gonna roast LLMs tuff!

aare they playing with each other ?

You can also play with LLms, btw

open source ?

yes!

i'll test it out . And will try making some good contributions.

That's very amazing! Excited to see what you contribute.

checkout:

This is amazing! 🔥

That's amazing! ❤️

Interesting... Eval fn's key, but long-term gameplay needs more than just next-move prediction.. behavioural modelling, maybe?

This is great!

Truly! You can also play with LLms, btw