HuatuoGPT-o1 🏥 a medical LLM designed for medical reasoning released by CUHK shenzhenModel:Code:Data:Paper:✨ 7B/8B/70B/72B8B&70B supports English

发布时间: 2024-12-31 17:50:27

1分

数据加载中

关注推特

收听电报

2

1

0

HuatuoGPT-o1 🏥 a medical LLM designed for medical reasoning released by CUHK shenzhen
Model:
Code:
Data:
Paper:
✨ 7B/8B/70B/72B
8B&70B supports English
时政
( twitter.com )

3个月前由 Adina Yakup 提交

HuatuoGPT-o1 🏥 a medical LLM designed for medical reasoning released by CUHK shenzhen
Model:
https://t.co/tZfFrjaa2A
Code: https://t.co/1nFI7t8cIk
Data: https://t.co/I0zQtg1s79
https://t.co/XE31aGBTKu
Paper: https://t.co/hm7hYwMsUN
✨ 7B/8B/70B/72B
8B&70B supports English
7B&72B supports English & Chinese
✨Excels with just 40K verifiable problems

Markdown支持

评论加载中...

您可能感兴趣的：更多

1

2

1

1

They showed med students what a c-section looks like 😭
有趣
( twitter.com)

8个月前 • non aesthetic things • -- 点击 • 下载视频 0 评论

00:00:26

2

2

1

1

They showed med students what a c-section looks like 😭
有趣
( twitter.com)

3个月前 • non aesthetic things • -- 点击 • 下载视频 0 评论

00:00:26

3

2

1

1

last night i demo’d o1 pro to my most skeptical friend - the results were extraordinary.
it was amazing to see o1 pro successfully tackle humanities research, and how much better it was than 4o / o1. i also had a ton of fun watching my friend come around to AI! story ⬇️
时政
( twitter.com)

3个月前 • Kay Sorin • -- 点击 0 评论

4

2

1

1

🍓Marco-o1! Newly Open-Sourced o1: Towards Large Reasoning Models for Open-Ended Solutions.
🎯 Built and released a CoT dataset to activate LLMs' reasoning abilities.
💡 Integrated LLMs with MCTS to expand the solution space.
🔬 Exploited action granularities in MCTS and
时政
( twitter.com)

2个月前 • Longyue Wang • -- 点击 0 评论

5

3

2

2

在 o1 pro 发布后，研究机构 SemiAnalysis 算是首次揭露了 OpenAI o1 系列的架构，还包含大量的推测和非公开渠道信息，值得一读。
我好奇的其实就是这几个问题： o1 系列和其他大语言模型有什么本质区别？ o1 和 o1 pro 之间的差异点在哪？ Tokenomics 是如何考量的？
1️⃣ o1 系列基于「Chain of
时政
( twitter.com)

3个月前 • Zhixiong Pan • -- 点击 0 评论

6

2

1

1

"Is 112345679 a prime number?" My friend asked me to test this on Grok 3 reasoning this morning. Unsurprisingly, Grok nailed it with sharp reasoning -- not a prime number as it’s 89 × 1262311! Btw, neither o1 nor o3-mini-high can answer it correctly.
🔗 Links below
- Grok-3
btc
( twitter.com)

1个月前 • Bill Yuchen Lin • -- 点击 0 评论

7

2

1

1

QwQ 32B is one of the models I’d want downloaded if I lost internet. Here it is streaming from my laptop to my phone at 15 tok/s.
This model is competitive with o1-mini in a few domains. Wild that you can serve it to your whole network from a laptop.
时政
( twitter.com)

3个月前 • Aaron Ng • -- 点击 • 下载视频 0 评论

00:00:26

8

2

1

1

OpenAI的o1新发布，不出意外就是去年年底传言的Q算法: Q learning和A路径搜索的组合，所以O1应该也算是一种GPT4和搜索的组合，意味着每次推理上算力消耗会大得多
o1给我感受最深的一点是
System1(快思考)的scaling law时代可能即将结束了，system2(慢思考)的scaling law新时代要来临了
IT技术
( twitter.com)

7个月前 • fin • -- 点击 0 评论

9

2

1

1

从截图还原网页，Claude sonnet 3.5 最强，o1 都比不上。
图一：Claude
图二：GPT-4o
图三：o1
图四：Gemini Flash 2.0
IT技术
( twitter.com)

3个月前 • 宝玉 • -- 点击 0 评论

0.03754 Second , Gzip Enable.本网所有言论均来自网络，不代表本网站立场。联系方式: [email protected]

©2012.11.21 bad.news All rights reserved. 社区自动运营第 -- 年零 -- 天
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

关注推特