on the model or your end developer's responsible use of the model itself. So we recently shared uh an anecdotal experience where we were skeptical of the benchmark saying that GLM was better than Opus. So we tested them on a real bug from the client repo and while both models fixed the issue, GLM was the winner in terms of cost and code quality. So GLM used twice as many tokens but only cost half as much. Opus finished faster. It used half as many tool calls. But GLM cleaned up dead code and verified that the build compiled before completing while Opus didn't. It left a bunch of type errors and it broke the production build. Um, and so that gave us the sense that GLM was trained to spend more tokens verifying its output. Um, which is fine because the tokens are cheaper any anyways and it's really the end result that matters.