Create accountLogin

Recent searches

No recent searches

Search options

Only available when logged in.
urbanists.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
We're a server for people who like bikes, transit, and walkable cities. Let's get to know each other!

Administered by:

Server stats:

523
active users

urbanists.social: AboutProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.3.8

#gemini25

0 posts0 participants0 posts today
Erik Jonker<p>"That is the nature of the Jagged Frontier. In some tasks, AI is unreliable. In others, it is superhuman"<br><a href="https://www.oneusefulthing.org/p/on-jagged-agi-o3-gemini-25-and-everything" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">oneusefulthing.org/p/on-jagged</span><span class="invisible">-agi-o3-gemini-25-and-everything</span></a><br><a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> <a href="https://mastodon.social/tags/AGI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AGI</span></a> <a href="https://mastodon.social/tags/Gemini25" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Gemini25</span></a> <a href="https://mastodon.social/tags/o3" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>o3</span></a></p>
Erik Jonker<p>What the "best" model is, really depends on your context and application. Benchmarks don't tell the whole story, are gamed etc. Still this benchmark about hallucinations is interesting because it shows that the one of the best current models, OpenAI o3 has a terrible score, 6,8% hallucination rate, quite terrible if you think about it. Gemini 2.5 is much better (1,1%).<br>Always check outcomes of any model...<br><a href="https://github.com/vectara/hallucination-leaderboard" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">github.com/vectara/hallucinati</span><span class="invisible">on-leaderboard</span></a><br><a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> <a href="https://mastodon.social/tags/Hallucinations" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Hallucinations</span></a> <a href="https://mastodon.social/tags/OpenAI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>OpenAI</span></a> <a href="https://mastodon.social/tags/o3" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>o3</span></a> <a href="https://mastodon.social/tags/Google" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Google</span></a> <a href="https://mastodon.social/tags/Gemini25" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Gemini25</span></a></p>
Erik Jonker<p>Interesting, i asked GPT-4o, o3, Gemini 2.5, Claude 3,7, at how many points lines crossed in the image below. GPT-4o said 5, o3 took 2 minutes, but gave the correct answer 8 and it used python code.<br>Gemini 2.5 answered quickly but failed , it answered 4.<br>Claude 3.7 also gave the correct answer and quickly.<br><a href="https://mastodon.social/tags/openai" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>openai</span></a> <a href="https://mastodon.social/tags/google" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>google</span></a> <a href="https://mastodon.social/tags/testing" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>testing</span></a> <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> <a href="https://mastodon.social/tags/o3" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>o3</span></a> <a href="https://mastodon.social/tags/gemini25" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>gemini25</span></a> <a href="https://mastodon.social/tags/claude" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>claude</span></a></p>
Erik Jonker<p>OpenAI o3 is far from AGI but especially because of it's web access when answering a query, it has become a very good assistant for answering various questions. It has a lot of added value for me. Gemini 2.5 is just as good but i don't have a paid account there so there are limits on queries.<br>Remember, never use private, confidential, company or organisational data (unless public) and always always check outcomes when you are going to use these for something else.<br><a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> <a href="https://mastodon.social/tags/o3" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>o3</span></a> <a href="https://mastodon.social/tags/Gemini25" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Gemini25</span></a></p>
PUPUWEB Blog<p>Google rolls out Gemini 2.5 Pro Experimental to all Gemini users, after initially launching it for Gemini Advanced subscribers on March 25. Exciting updates ahead! 🚀 <a href="https://mastodon.social/tags/Google" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Google</span></a> <a href="https://mastodon.social/tags/Gemini" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Gemini</span></a> <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> <a href="https://mastodon.social/tags/TechNews" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>TechNews</span></a> <a href="https://mastodon.social/tags/Innovation" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Innovation</span></a> <a href="https://mastodon.social/tags/MachineLearning" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>MachineLearning</span></a> <a href="https://mastodon.social/tags/AIUpdates" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AIUpdates</span></a> <a href="https://mastodon.social/tags/Gemini25" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Gemini25</span></a></p>
Erik Jonker<p>Gemini 2.5 solved this connections puzzle flawlessly, impressive.ChatGPT 4o failed !<br><a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> <a href="https://mastodon.social/tags/NYT" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>NYT</span></a> <a href="https://mastodon.social/tags/connections" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>connections</span></a> <a href="https://mastodon.social/tags/gemini25" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>gemini25</span></a> <a href="https://mastodon.social/tags/chatgpt" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>chatgpt</span></a></p>
Erik Jonker<p>Good analysis by "AI explained"<br><a href="https://youtu.be/Y9mVlNwj_ic?si=iRWg8lUWyTgc11BP" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">youtu.be/Y9mVlNwj_ic?si=iRWg8l</span><span class="invisible">UWyTgc11BP</span></a><br><a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>AI</span></a> <a href="https://mastodon.social/tags/Gemini25" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Gemini25</span></a> <a href="https://mastodon.social/tags/Deepseek" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Deepseek</span></a> <a href="https://mastodon.social/tags/Microsoft" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>Microsoft</span></a> <a href="https://mastodon.social/tags/OpenAI" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>OpenAI</span></a></p>
ExploreLive feeds
About