- 主题:真正的开源LLM来了，OLMo 2 32B发布 (转载)
【 以下文字转载自 NewExpress 讨论区 】
"OLMo 2 32B: First fully open model to outperform GPT 3.5 and GPT 4o mini"
"OLMo is a fully open model: [they] release all artifacts. Training code, pre- & post-train data, model weights, and a recipe on how to reproduce it yourself."
--
FROM 182.118.238.*
赞，这才是真正的开源
【 在 upndown 的大作中提到: 】
: "OLMo 2 32B: First fully open model to outperform GPT 3.5 and GPT 4o mini"
: "OLMo is a fully open model: [they] release all artifacts. Training code, pre- & post-train data, model weights, and a recipe on how to reproduce it yourself."
: --
: ...................
--
FROM 202.47.106.*
如果为真，那可就太牛逼了
【 在 upndown (每天锻炼2小时为祖国健康工作100年) 的大作中提到: 】
: 【 以下文字转载自 NewExpress 讨论区 】
: "OLMo 2 32B: First fully open model to outperform GPT 3.5 and GPT 4o mini"
:
: "OLMo is a fully open model: [they] release all artifacts. Training code, pre- & post-train data, model weights, and a recipe on how to reproduce it yourself."
--
FROM 222.128.172.*
看了一下 性能不够 没有兴趣
现在的AI开源的只有前三才有可能被大规模铺开
【 在 upndown 的大作中提到: 】
: "OLMo 2 32B: First fully open model to outperform GPT 3.5 and GPT 4o mini"
: "OLMo is a fully open model: [they] release all artifacts. Training code, pre- & post-train data, model weights, and a recipe on how to reproduce it yourself."
: --
: ...................
--
FROM 125.120.165.*