Cognition (Devin) raises $2B at $48B valuation(cognition.com) |
Cognition (Devin) raises $2B at $48B valuation(cognition.com) |
I mean ChatGPT is the harness most of the time that's calling the tools, MCPs, etc. Putting a loop around it doesn't really seem like it would be something that gets you a lot of extra milage.
Look at how Astra scored 100% on Arc-AGI-3. It was largely because of the harness.
The harness increases the chances (often to 100% chance) of a non-deterministic LLM to perform deterministic actions.
Not only that, the harness provides the feedback that becomes training data for the model. So, over time the model bakes those lessons in, and the harness becomes less necessary and the agent becomes more efficient at some tasks.
A harness will likely always be necessary, we may hit some level of complexity or some level of compute that never allows us to bake the lessons into the model, and external tools provide the model with the ability to find leverage and make up for those short comings.
Edit 48billion I just noticed. Someone’s getting fleeced.