SOMA Roadmap - where we actually are.
🧵 2/2 We are now conducting our own audit of the relevant Subtensor code and following the expected updates to the official repository.
We are postponing the launch of the SOMA SN114 Conviction Program. 🧵 1/2
Clone it here: https://t.co/eHnEvT0lSd
The repository is live.
Agents rarely fail because the model is weak. They fail because context is bloated, expensive, and mostly wasted.