Comment by redox99
2 days ago
Yep. These small models are actually worse than GPT 3.5 at some tasks (like recalling facts). You can definitely make models smarter at specific tasks (like tool calling, coding) but you can't compress the entire human knowledge into a 30GB file. It's just not enough bits.
But why would you use a model to store factual knowledge, that is stupid. We want intelligence, not a database.
Models cant make any decisions if they have 0 idea that the feature exist in this language or in some general fact.
For example you would tell a model hey, become an expert in this language for me, search it online, it would still need to learn it and download the data to it's context and then increasing the memory usage, there's no way around it.
Implicit knowledge is different from remembering every church that exists in the state of NSW Australia
That's a good sign.
Ideal local model would not know stuff like who Britney Spears is, best to leave precious weights for something useful.
Of course the line is very blurry but I'd be perfectly happy with local model that doesn't know anything about history, geopolitics, art or even biology etc. just coding, operating systems etc.
At one point there were specific -Coder release of Qwen e.g. 2.5 but they dropped that, still wondering how much better 3.6-Coder or 3.8-Coder would be when they ignore everything else
that is why we enable web search for the agent. the memory can come from the internet.
deepseek-v4-flash needs web search to return true facts.
Yeah but results are worse, see https://news.ycombinator.com/item?id=49301574
In my experience, the facts that are compressed away in small models are ones you don’t need them to memorize. They need familiarity with the essential concepts in a field, so that they will have better comprehension of material put into the context (and make better retrieval decisions). They don’t need to know very particular details which are a hallucination risk and should be derived from the context instead.