michabbb<p>Recent breakthroughs like <a href="https://social.vivaldi.net/tags/OlympicCoder" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OlympicCoder</span></a> outperform <a href="https://social.vivaldi.net/tags/Claude3" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Claude3</span></a>.7 on coding tasks with just 7B parameters, while <a href="https://social.vivaldi.net/tags/AI2" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI2</span></a>'s <a href="https://social.vivaldi.net/tags/OLMo2" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OLMo2</span></a> models match <a href="https://social.vivaldi.net/tags/OpenAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OpenAI</span></a>'s o1-mini performance.</p>
mastodon.online is one of the many independent Mastodon servers you can use to participate in the fediverse.
A newer server operated by the Mastodon gGmbH non-profit
Administered by:
Server stats:
11Kactive users
mastodon.online: About · Status · Profiles directory · Privacy policy
Mastodon: About · Get the app · Keyboard shortcuts · View source code · v4.4.0-nightly.2025-03-25
#olmo2
0 posts · 0 participants · 0 posts today
Hacker News<p>Mlx-community/OLMo-2-0325-32B-Instruct-4bit</p><p><a href="https://simonwillison.net/2025/Mar/16/olmo2/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">simonwillison.net/2025/Mar/16/</span><span class="invisible">olmo2/</span></a></p><p><a href="https://mastodon.social/tags/HackerNews" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>HackerNews</span></a> <a href="https://mastodon.social/tags/MlxCommunity" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MlxCommunity</span></a> <a href="https://mastodon.social/tags/OLMo2" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OLMo2</span></a> <a href="https://mastodon.social/tags/4bit" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>4bit</span></a> <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://mastodon.social/tags/Technology" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Technology</span></a> <a href="https://mastodon.social/tags/Innovation" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Innovation</span></a></p>
michabbb<p><a href="https://social.vivaldi.net/tags/OLMo2" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OLMo2</span></a> 32B: First fully <a href="https://social.vivaldi.net/tags/opensource" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>opensource</span></a> model to outperform <a href="https://social.vivaldi.net/tags/GPT3" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>GPT3</span></a>.5 and <a href="https://social.vivaldi.net/tags/GPT4o" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>GPT4o</span></a> mini 🔥</p><p>🧵👇 <a href="https://social.vivaldi.net/tags/MachineLearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MachineLearning</span></a> <a href="https://social.vivaldi.net/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://social.vivaldi.net/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a></p>
BedastGPT<p>I’m playing with AllenAI OLMo2 LLM.</p><p>It kept insisting my use of insulin and my TDD indicated high insulin resistance.</p><p>When I noted my c-peptide result is around 1.1ng/dL, it adjusted its view and indicated my TDD would present a challenge in keeping up with goals as a T2D.</p><p>My Time In Range is well over 90% while using an insulin pump with a TDD between 30 and 40 units per day.</p><p>When I talk about how T2D can be complicated, this is what I mean. Things aren’t always as simple as “lose weight, cure diabetes”.</p><p>And this is why we put our trust in humans who are able to reason with modern data to provide our healthcare.</p><p>I will say, among the small LLMs that you can run on your own at home, OLMo2 is among the most impressive in terms of emulating knowledge. It’s wild that it doesn’t get more attention. I have to keep reminding myself that it’s very much confidently lying to me.</p><p><a href="https://beige.party/tags/Diabetes" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Diabetes</span></a> <a href="https://beige.party/tags/T2D" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>T2D</span></a> <a href="https://beige.party/tags/Type2Diabetes" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Type2Diabetes</span></a> <a href="https://beige.party/tags/AllenAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AllenAI</span></a> <a href="https://beige.party/tags/OLMo2" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OLMo2</span></a> <a href="https://beige.party/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a></p>
ExploreLive feeds
Mastodon is the best way to keep up with what's happening.
Follow anyone across the fediverse and see it all in chronological order. No algorithms, ads, or clickbait in sight.
LoginDrag & drop to upload