Using Chat with RTX

After training Chat with RTX on our 60,000 news articles, we started quizzing it. None of the questions listed were intentionally crafted to be difficult for AI. We simply wondered "what could we ask this thing?." We began by asking what's the amount of cache on Raptor Lake—a rather nebulous question, once you think about it, since we didn't mention the processor model. It put out the answer 68 MB, with 36 MB of L3 cache, and 32 MB of L2 cache—the correct answer (8x 2 MB L2 for the P-cores, and 4x 4 MB for the E-core clusters in the 8P+16E Raptor Lake-S silicon). It ends its answer with a link to the text file it drew this information from. It's obviously not copy pasting statements from the news post, but composing a natural language answer based on the information it sees.


Here's another example, where we ask it how a GeForce RTX 4080 SUPER differs from the original RTX 4080 from 2022. Most information is correct, but the TGP is incorrectly listed as 450 W. Our original article is very clear that this is the theoretical limit of the cable configuration, not the actual power limit.


These answers make very little sense to the educated reader, even though they look plausible at a first look—one of the biggest dangers of AI-generated texts.


In its current state, Chat with RTX isn't able to connect follow-up questions to an original question. This is mentioned in NVIDIA's reviewer's guide, so it's not unexpected.


Next, we started feeding it some queries that we might encounter in real-life, and here things aren't looking so good.




Some of the answers are plainly wrong. But here, too, they do look very credible at first look.

Interpreting YouTube Videos?

An interesting feature with the Chat with RTX application is its ability to interpret and answer questions related to YouTube videos. You feed in a YouTube video URL, or the link to a playlist, it indexes the data and can answer questions. Very impressive!


Here's how it really works—when you point it to a YouTube video, it looks for the English language closed captions (CC) file, which is provided by YouTube. There's no magic happening here, the application isn't interpreting the audio/video stream of the video, no image recognition or speech to text happening at all—it's simply answering questions based on the CC file it downloads from YT. No doubt that's a fantastic idea and will make it much easier to condense the information in long videos. In some queries you'll notice that the responses even include SDH cues like "[Music]."