Google Gemini vs OpenAI ChatGPT

[admin]

In a battle of the chatbots I’ve put Google’s Gemini up against OpenAI’s ChatGPT to see which performs best on a series of tests.

Google Gemini vs OpenAI ChatGPT : Read more

 

[czorrilla]

I tested the free version of both last Friday. I needed to create a job description, and to do it faster I used another one as template. Gemini gave me the better response. I only had to remove a few lines and be done.

 

[RyanMorrison]

I tested the free version of both last Friday. I needed to create a job description, and to do it faster I used another one as template. Gemini gave me the better response. I only had to remove a few lines and be done.

That’s interesting. I have found that Gemini is more consistently responsive.
I’ll be testing the pro versions over the next few weeks

 

[nsteblay]

I use LLMs for business applications. There really is no comparison across all the models at this time. ChatGPT is by far the most consistent.

 

[hundvd7]

I have been using ChatGPT for the past few months, and I've been trying to move to Bard for a month or so, but it fails me really often.

One weird quirk it has that it absolutely refuses to translate anything.

I've tried many many times, and with different languages, and I get "I am am a language model, I am not capable of that" every single time without fail.

 

[RyanMorrison]

I have been using ChatGPT for the past few months, and I've been trying to move to Bard for a month or so, but it fails me really often.

One weird quirk it has that it absolutely refuses to translate anything.

I've tried many many times, and with different languages, and I get "I am am a language model, I am not capable of that" every single time without fail.

Gemini does have a big refusal problem, particularly for things that are region specific like translation and image creation. This should go away as Google rolls it out more widely.

 

[c13mw]

Will there be another full comparison article published comparing the paid OpenAI GPT-4 vs Gemini Advanced? In the article you mentioned that you fed responses into Claude 2, Mixtral, Gemini Advanced and ChatGPT Plus. It'll be awesome to see the 9-round face-off differences compared to the free versions.

 

[Nintendaz]

Would love to see the premium version comparisons. I have both premium and I usually go to ChatGPT when I need something very personalized or specific and I go to Gemini when I need something a bit more up to date or quick.

I'm currently using both mainly to help me learn UE5 and implement features for a game but both have very outdated knowledge (neither use 5.1+ features in their implementation). They both are inconsistent at understanding the best way to implement features though. I will give a plus to ChatGPT as if i just screenshot my blueprints and send it to it, 19/20 times it'll understand whats going on whereas Gemini doesn't seem to understand the image half the time.

I do prefer the way Gemini responds overall as if i request help with multiple things it wont try to cram everything into a single prompt and instead it breaks it down but remembers what other things ive wanted to do. However I can easily give this structure to ChatGPT as well.

I need to compare them on more things but I personally prefer ChatGPT only because I can personalize it so much more. I've used ChatGPT for meal planning, workout routines and more and it nails every single one of these more "simple" requests.

Currently making training data to feed it with updated UE5.3 info and if it goes well, I doubt I'll be going back to Gemini anytime soon. I also doubt I'll use Gemini after I get access to the memory feature in ChatGPT, but we'll see.

Edit: Also forgot to mention, the extensions feature with Gemini kinda sucks right now. Only use case I've had so far is summarizing YouTube videos. There's so much potential for it though so I'm hopeful.

 

[RyanMorrison]

Will there be another full comparison article published comparing the paid OpenAI GPT-4 vs Gemini Advanced? In the article you mentioned that you fed responses into Claude 2, Mixtral, Gemini Advanced and ChatGPT Plus. It'll be awesome to see the 9-round face-off differences compared to the free versions.

Yes. That will be coming in the next couple of weeks. I've already started working on it.
I will also do a comparison with Claude 3 soon, now that the new model has launched and claims to beat GPT-4 in key benchmarks.

 

[RyanMorrison]

Would love to see the premium version comparisons. I have both premium and I usually go to ChatGPT when I need something very personalized or specific and I go to Gemini when I need something a bit more up to date or quick.

I'm currently using both mainly to help me learn UE5 and implement features for a game but both have very outdated knowledge (neither use 5.1+ features in their implementation). They both are inconsistent at understanding the best way to implement features though. I will give a plus to ChatGPT as if i just screenshot my blueprints and send it to it, 19/20 times it'll understand whats going on whereas Gemini doesn't seem to understand the image half the time.

I do prefer the way Gemini responds overall as if i request help with multiple things it wont try to cram everything into a single prompt and instead it breaks it down but remembers what other things ive wanted to do. However I can easily give this structure to ChatGPT as well.

I need to compare them on more things but I personally prefer ChatGPT only because I can personalize it so much more. I've used ChatGPT for meal planning, workout routines and more and it nails every single one of these more "simple" requests.

Currently making training data to feed it with updated UE5.3 info and if it goes well, I doubt I'll be going back to Gemini anytime soon. I also doubt I'll use Gemini after I get access to the memory feature in ChatGPT, but we'll see.

Edit: Also forgot to mention, the extensions feature with Gemini kinda suck right now. Only use case I've had so far is summarizing YouTube videos. There's so much potential for it though so I'm hopeful.

I'm working on the premium comparison now — hopefully go live in a couple of weeks.

One of the big advantages of ChatGPT is being able to set custom instructions that remain across all chats, it is particularly useful as you say for things like meal planning.

I'm also reviewing Claude 3 Pro to see how well it works. I'll give it a try with some ULE5 blueprints and see how it works. Its knowlede cut off is August 2023.

 

[nellea]

I have tried both for my code projects and i must say OpenAI's ChatGPT is 100 times better than Google Gemini. Gemini didnt give me the desired output that i want and only chatgpt is capable for that.