Small language models tuned narrowly for a specific task (like allow/deny enforcement decisions) outperform generically prompt-engineered large frontier models on that task.

23:48plug for Rubrik, which is actually doing this. But I do think that, you know, to your point, the the core is like, can you find a system that basically architects both of these things together? Cuz you probably have something for data backup and recovery for like other compliance reason or other. Is it possible to connect it into your observability? And I will actually say one misnomer is that the SLM is not

24:08as good as at enforcement on the LLM. Actually, we find that for domain-specific tasks, small language models that are tuned for a very specific task tend to outperform a generically prompt-engineered larger language model. So, like when we benchmark our SLM versus at the time, I think we benchmarked to GPT-5 2 as an example, we found that not only were we an order of magnitude faster and cheaper, but we

24:28were also actually more accurate on being able to make a binary classification on whether or not to allow or disallow. And we've relatively consistently seen that fact when you just constrain the outputs of the LLM to be very low cardinality in terms of what it's supposed to do, which is exactly what you want from a guardian agent. >> Yeah, I could see that. I think what colored my perspective on that is

24:49thinking about with Open Claw or, you know, personal agent, it is frequently said that you should use like a frontier model for your main orchestrator because your SLMs, you know, if you expose them to external untrusted data sources, they can be easily manipulated relative to a frontier model that, you know, A is built tuned better to to be able to detect and resist manipulation, uh but also it's just like smarter, more

25:20parameters. >> So, we've done a we've done a lot of benchmarking on this exact line of thinking and uh did it even at Protobase. Like we released a paper called Loreland because all of these are like Laura tuned adapters essentially uh as a technical detail. But uh what we found was that for open-ended like for open domain tasks, you're 100% right. Use a large frontier model. That's why the orchestrator or

25:40the planner should really be like a larger model. But if the more constrained you get into tasks, the better and better you actually tend to see performance from small >> a fixed domain classifier, then >> If it's a fixed domain classifier, the best you could usually do is post-training with like SFT, um you know, a small model on it. And then if you have a task like what we're talking

26:02about, which is simply should this request be permitted or denied, like that is actually the ideal type of task for an SLM. Um and that's where you're able to run it at super low latency. Now, I think to your point, like ultimately, I think that uh all prevention mechanisms are going to have some rate of false negatives. Um and like the agent world is just moving too quickly to be able to catch