You should work on more and better benchmarks before claiming it's value. The benchmarks present are naive ones and only count tokens. What about performance of the model? Is it really better than caveman or just simple 'be brief'? Only numbers could tell
Have you studied your plugin's impact on the model's effective intelligence? Such plugins are a dime a dozen now and we need to get past touting token savings. You can save more if you just throw everything away!
You should work on more and better benchmarks before claiming it's value. The benchmarks present are naive ones and only count tokens. What about performance of the model? Is it really better than caveman or just simple 'be brief'? Only numbers could tell
Have you studied your plugin's impact on the model's effective intelligence? Such plugins are a dime a dozen now and we need to get past touting token savings. You can save more if you just throw everything away!
[flagged]
How does this differ from context-mode plugin?
[flagged]
[flagged]
[flagged]
[dead]