>Tested against a Java monorepo across four scenarios, measuring tokens Claude would consume reading files directly vs. consuming the bulk-reader's summary or writing code via the code-writer. Mean bulk-read savings were around a whopping 90%.
>The code-write scenario is harder to measure in tokens because without shunt, Claude both reads the reference files and generates the output as expensive output tokens. With shunt, the code goes straight to disk, Claude never sees it.
So nothing about accuracy or actual performance? At least run against DeepSWE bench or something.
It cuts token usage because they are using a different service with a different token budget for the reader/code writer tasks.
You can also just delegate this to subagents with Claude Code (though you have a more limited choice of models unless you swap the cheaper models via OpenRouter).
I'm OK using a dumb model as a smart grep, but the whole point of using the frontier models is using their intelligence for the hard stuff like coding.
Isn't this a somewhat standard multi-model setup? there's nothing ground breaking here, just delegate claude to plan -> smaller model for implementation.
>The benchmarks
>Tested against a Java monorepo across four scenarios, measuring tokens Claude would consume reading files directly vs. consuming the bulk-reader's summary or writing code via the code-writer. Mean bulk-read savings were around a whopping 90%.
>The code-write scenario is harder to measure in tokens because without shunt, Claude both reads the reference files and generates the output as expensive output tokens. With shunt, the code goes straight to disk, Claude never sees it.
So nothing about accuracy or actual performance? At least run against DeepSWE bench or something.
So this is just delegating certain work to dumber models? I certainly wouldn't use Gemini 2.5 Flash (!!?) for code writing as suggested.
I've never had an issue with Codex or Claude reading massive files, they're really good at precise greps.
Yes, this makes little sense. It looks like it's a way to avoid having Claude read or write your code.
And why stop at 90%? I have this one weird trick to reduce Claude Code token use by 100%: use a different harness and model!
It cuts token usage because they are using a different service with a different token budget for the reader/code writer tasks.
You can also just delegate this to subagents with Claude Code (though you have a more limited choice of models unless you swap the cheaper models via OpenRouter).
I'm OK using a dumb model as a smart grep, but the whole point of using the frontier models is using their intelligence for the hard stuff like coding.
Isn't this a somewhat standard multi-model setup? there's nothing ground breaking here, just delegate claude to plan -> smaller model for implementation.
This is just offshoring but for models