My Experience with the Lam Playground on the r1 Device

Overview

From my initial attempts to use the Lam Playground to now, I have noticed significant improvements. I believe this is due to both my growing familiarity with the interface and enhancements in the overall performance of the tool.

Challenges Faced

  • Credential Setup: I had to set up new credentials in the cookie jar after the

Subject: Challenges with Forum Post Creation on the r1 Device

Content:

I recently encountered some challenges while creating a forum post using the r1 device. After creating a new topic, I found it difficult to have the complete text read out by the device. When I requested it to read the topic, it took a considerable amount o

Overall, it took about 40 minutes to create 4 different posts in the forum. There were several hiccups still present, which already shows how effective this is.

@simon @rabbit @mattdomko

Another example of only partly successful attempts …

Yeah, this happened often to me as well. Actually, it’s too limited; I think in the future they’ll charge for this service… it sad to say but we’re beta testers for now here.

To work around this, I’m creating a draft inside my Gmail account (already generated with Copilot or Perplexity) and I tell my R1 to take this draft to post. Anyway, I’m experiencing this very often, and I’ve understood that R1 is unable to copy and paste but it reads and later rewrites the message. So it would be great to get information from the Rabbit Inc. Team here about how many characters are permitted for our R1 because it’s really limited.

Probably they will definitely implement a paid model because all this token generation comes at a cost! …and not a small one. That’s why in another post I’m encouraging consideration of interaction with macro “repetitive operations” that our dear little rabbit could launch without wasting tokens. But I understand it’s difficult to both comprehend and implement this concept.

I suggest you vote for this feature :wink:

…as suggested also from @iRockxul (Macro Creation) and someone else…

Don’t know. I like their vision of the pay once model, and really hope they can stick with it. So far, it’s been great, and I really like the stubbornness of @jesse and his aversion to subscription models.

We’ll see, I guess things will improve over time, and hopefully not too far away, it will work properly.

BTW: I don’t think that these input sudden stops are a cause of token limit. I think it is more that the model is hanging somewhere, or has problems inputting text.

As just writing that it took about 40 minutes to create 4 different forum posts is hard to follow, I took a video now that demonstrates some of the problems: