Google just launched a new model on March 26th which they claim to be the best on coding, reasoning and overall everything. 🥴 But I mostly care on ...
For further actions, you may consider blocking this person and/or reporting abuse
Love the short, sweet intro to Gemini 2.5. Yes, this is a beast of a model.
Agree 100% on the Gemini 2.5. Haven't tried out Gemma.
Good to hear that. BTW, if you want to try Gemma 3 out locally, you might find this repository of mine that helps to set up LLMs on a VM helpful.
Thank you. I don't use it locally. I use it in the AI studio.
Surely, that could be the case. Gemini 2.5 performed quite well in these tests. I haven't really tested it on the UI side with Tailwind and all that, but I can't agree more on how good Claude 3.5/7 is with backend stuffs. It's awesome. Thank you, Kevin! I'm glad you took the time to read this one!
Stunning results. It's really hard to catch up with constant updates of all the LLMs.
Just experimented with the improved image generation in ChatGPT.
Thanks again for sharing your results. I really like the flight sim one by Gemini2.5!
Completely understand that with so many LLMs, it's hard to keep up with the updates.
And thank you for checking it out, Benny! 🔥
I really appreciate how you explained all the information smartly. We are waiting for your further blogs. We are also looking for information about aquarius and taurus compatibility to clarify how gemstones can have specific effects on our lives. We request you to please visit our website and give suggestions and feedback. And the wait continues for your coming blog.
Thank you for this comparison! I recently got to know this model from Mathew Berman and really excited to try this out in my day-to-day workflow. Good job with the comparison! 🔥🫶
Glad you enjoyed!
🥰
Good model comparison. 💯
Thank you, @shekharrr 🙌
How to use gemini 2.5? I don't find it anywhere. Still use GPT
You can find it in the Google AI Studio: aistudio.google.com
Aago sathi! Bholi exam nabigara hai tara 😂
Thank you! Bigrinna :)
Always love your comparison, Shrijal. 👍🏻
You seem to be a go-to nowadays for AI models comparison. Love it!!
How do you like the new DeepSeek v3?
That means a lot. Thank you, Lara!✌️
I haven't really tried it yet, but I will soon, and I'll share my thoughts with you
upload it a bit of context and it does not. I had the issue with plantuml diagrams and threw it 200K tokens of pdf documentation on the context as pdf and kaboom: most problems are gone
Guys, do let me know your thoughts in the comments! ✌️
You can also find this blog here: Link
Thank you.
Thank you for the feedback on the models. 👍
Thank you! 🙌
This is awesome!