Google's New Gemini 0121: Smarter Than DeepSeek R1? (Reasoning & Coding Test!)

[@Codewello]

What do you think about this model ? 🤔

[@alexpol108]

Thanks for the quality content, waiting for the video you mentioned!

[@Codewello]

I am glad you enjoyed it, I'll try to have it out as soon as possible! 😄

[@Spandy0]

Tbh For coding I still prefer qwen and DeepSeek.

[@kamalikapoddar]

How are you using deepseek my APIs don’t work at all

[@SnazzieTV]

@@kamalikapoddar probably running locally. Can use something like LM Studio

[@danial_amini]

​@@kamalikapoddarthe deepseek api page won't even open for me

[@CuentaUno-ls8oi]

​@@kamalikapoddarThe system of Deepseek in China has gone down for many hours due to poor infrastructure

[@CuentaUno-ls8oi]

Claude 3.5 artifacts is better for coding, Qwen cannot generate graphs for research still

[@radstartrek]

Its blazing Fast and pretty good so far.

[@Codewello]

yes but the rate limiting is driving me crazy

[@Todestelzer]

o3 mini high seems to be the best model for coding right now. But you need to have bought 250$ in API tokens to get access. Really impressive how good AI is getting and how fast it improves.

[@Codewello]

I was really wondering how I was going to test this O3-High model, and now I have the answer. Thanks for the info!

[@gemini_537]

This is the best model in the LM Arena!!

[@Codewello]

for now

[@lowdmedia]

This updated model is awesome. The rate limits really cripple it though.

[@Codewello]

Yeah, I feel you.

[@SeeFoodDie]

Fantastic. However Google needs flashier names for its models 😊

[@Codewello]

and shorter names ,The model names are too long.

[@UNCLESPARK]

I have used google Gemini for coding and I have the same issue….rate limits. Don’t get me wrong it is a great model for coding but over large data bases rate limits really mess it up

[@Codewello]

That is a known issue with most of the current new models from google

[@ayo4008]

lol assuming a c d are also males, the total number of males is 5. final answer is 3

[@heyitsronok]

not good

[@CuentaUno-ls8oi]

Is very good

[@Codewello]

it's in the top 5