How do you even use o1 as an agent? It’s gonna be a while before the tools integrate with it - ThreadSky

About ThreadSky

cfdevelop.bsky.social • 81 days ago

How do you even use o1 as an agent?

It’s gonna be a while before the tools integrate with it

Comments

Hard to have any real discussion on bsky because of the crappy character limits. At a high level, an agent needs to be able to come up with plans, then call functions while executing the plan. The 4o models can be used for some of those parts, but are rather low level and brittle for building agents

cfdevelop.bsky.social•81 days ago

I’ve used ChatGPT to do this and it’s pretty mediocre. It’s one step forward, one step back

But Claude often nails it with Cursor guiding it

jezell.bsky.social•81 days ago

The o1 models appear to basically just be refined versions of the 4o model that help avoid bad plans and package up that for loop into something that works out of the box with a lot less work. You can't even have a system prompt (which is needed for 4o agents), because they use it for their loop.

cfdevelop.bsky.social•81 days ago

Yes, from what I understand, o1 is just looping and checking its own output to improve accuracy

But that’s not really relevant with an agent loop because that’s what happens anyway.

Hallucinations get ironed out because the agent always returns to the goal

Posting Rules

Be respectful to others
No spam or self-promotion
Stay on topic
Follow Bluesky's terms of service

Comments

Posting Rules

Reply