Viewing profile — ivzak

Speaking from experience - serving good context compression is not trivial.

Thanks, checking it out!

We keep preserved pieces of context unchanged = compression removes some pieces of the input while keeping the others verbatim. Let us shortly share a concrete example

For auto-compact, we do essentially the same Anthropic does, but at 85% filled context window. Then, when the window is 100% filled, we pull this precompaction + append accumulated…

It seems to be the hit rate of a very straightforward (literal matching) retrieval. Just checked the benchmark description ( https://huggingface.co/datasets/openai/mrcr ), here it …

Probably LLM-generated, but that's a fair point :D Well, the proxy is open source, maybe someone will even implement this before we do :) Talking about the features proxy unlocks -…

Claude code still has /compact taking ages - and it is a relatively easy fix. Doing proactive compression the right way is much tougher. For now, they seem to bet on subagents solv…

I think we should draw distinction between two compression "stages" 1. Tool output compression: vanilla claude code doesn't do it at all and just dumps the entire tool outputs, blo…

Subagents do summarization - usually with the cheaper models like Haiku. Summarizing tool outputs doesn't work well because of the information loss: https://arxiv.org/pdf/2508.2143…

I doubt Anthropic would single-handedly cut their API revenue in half by rolling out compression. Zero incentive.

You’re right - poor compression can cause that. But skipping compression altogether is also risky: once context gets too large, models can fail to use it properly even if the neede…

Your comment reminded me of this old simulacra paper ( https://arxiv.org/pdf/2304.03442 ) :) iirc, they compressed the "memory roll" of the agents every once in a while

We don't sell data to model distillers.

There is a nice JetBrains paper showing that summarization "works" as well as observation masking: https://arxiv.org/pdf/2508.21433 . In other words, summarization doesn't work wel…

We built an open-source proxy that sits between coding agents (Claude Code, OpenClaw, etc.) and the LLM, compressing tool outputs before they enter the context window. Demo: https:…