Comment by @brianli • Hey

How should we be thinking about the timeline for sufficient compute on L1s or L2s to make this economically feasible?