Samsung released a 7M model that achieved 45% on ARC-AGI-1
1–10 of 14 posts
Re: Samsung released a 7M model that achieved 45% on ARC-AGI-1
#2Wow this is legitimately nuts
Re: Samsung released a 7M model that achieved 45% on ARC-AGI-1
#3Wow this is legitimately nuts
Why?
Re: Samsung released a 7M model that achieved 45% on ARC-AGI-1
#4Discussed here:
https://news.ycombinator.com/item?id=45506268 Less is more: Recursive reasoning with tiny networks (54 comments)
Re: Samsung released a 7M model that achieved 45% on ARC-AGI-1
#5related/duplicate https://news.ycombinator.com/item?id=45506268 I think.
Re: Samsung released a 7M model that achieved 45% on ARC-AGI-1
#6seems like they just stole the original HRM (just glanced at this though)
Re: Samsung released a 7M model that achieved 45% on ARC-AGI-1
#7Wow this is legitimately nuts
Why?
ARC-AGI is one of the few tests on which human can complete easily while LLMs still struggle. This model scores 45% on ARC-AGI-1 and 8% on ARC-AGI-2, the latter is comparable to Claude Opus 4 and GPT-5 High, behind only Claude Sonnet 4.5 and Grok 4 Thinking, for a model about 0.001% the size of commercial models.
Re: Samsung released a 7M model that achieved 45% on ARC-AGI-1
#8Released where?
Re: Samsung released a 7M model that achieved 45% on ARC-AGI-1
#9Can someone elaborate on the meaning of "7m model"?
I'm new to AI, and had an LLM spit out an explanation of why some of the "local" models don't work in Ollama on my Air, but... I don't know how accurate the AI is, heh.
It's my understanding most models are more like 1-30b (as in Billion)
Re: Samsung released a 7M model that achieved 45% on ARC-AGI-1
#10Can someone elaborate on the meaning of "7m model"? I'm new to AI, and had an LLM spit out an explanation of why some of the "local" models don't work in Ollama on my Air, but... I don't know how accurate the AI is, heh. It's my understanding most models are more like 1-30b (as in Billion)
7 million parameters