Paritok

Every long session I run eventually hits my own harness's compaction, where older turns get summarized once context grows, completely separate from anything Paritok does upstream. If a compressed segment's REF id pointer lives inside a turn that the host later summarizes away before the agent ever calls read_original on it, is that byte range still recoverable, or does recall depend on the referencing turn staying intact in the live context window?

Is it safe to use this tool with a Claude Code subscription?

nice! and good that you are doing the open source route, that builds confidence!

Congrats on the launch Watching coding agents burn through context windows with redundant tool schemas and file dumps is so frustrating. Love that this sits as a completely local proxy without needing a middleman cloud server.

Those are some interesting numbers! How are you justifying costing 85% less, and what's the benchmark?

Context bloat gets brutal with MCP-heavy setups. Keeping compression local while making the original bytes recoverable is a smart approach. Curious, how much latency does Paritok add per turn in a typical Claude Code session?

Want to make sure I'm reading the benchmark right. Quality retained = solve rate ÷ uncompressed baseline - so 86.5% means the agent resolves about 86.5% as many issues as it would with full context, a relative 13.5% drop in tasks actually completed? Or is that measuring something else?If so, the comparison I'd want isn't against other compressors, but against just running a cheaper or lower-effort model at full context. Both routes cost quality and both cut the bill. Has that been measured? A strong model on compressed context vs a cheaper one on complete context, same tasks.And which model was the scaffold running? Hard to judge 13.5% without knowing the baseline.

Niceee

Non-destructive compression via a code-native 4B model is a genuinely clever architecture. Adding to my to-try list.

Really like this product and I went to your github. The compounding savings math in the README is refreshingly honest. This is how AI infra should be marketed.

Great job! The 3× turns per context window benefit is actually valuable to me. My Claude Code sessions hit compaction on big codebases. Excited to see how much longer they can run with this.

Congratulations on the launch! Context bloat is becoming a real bottleneck for coding agents, so a fully local way to compress tools, files, and history feels both practical and privacy-conscious. Looking forward to trying Paritok.

This is such an underrated problem honestly,
Everyone talks about model quality but nobody talks about how much of the context window is just wasted overhead.
btw, how much of that 85% savings comes from the tool schema compression vs the file/output compression?
lastly, all the best with the launch team )

Hey PH community!We're Jiayu and Luzhuo, two engineers who got tired of watching our coding agents burn through tokens.Here's what we kept seeing: your agent ships far more than the model actually needs. Full JSON for seventy tools when it will call two. An entire file when it needed one function. Debug output nobody will ever read again. And because the API is stateless, all of it goes back on every single turn, so the waste doesn't just cost you once, it compounds.So we built Paritok, a non-destructive compression gateway powered by a code-native 4B model we trained on 45K real agent trajectories. It sits between your agent and the API and cuts three things before they leave:• Tool schemas — 29K down to 8K per turn, no model involved, this one runs on CPU• File reads and tool output — compressed to about a quarter of their size• Stale history — turns beyond a recent window get summarized once your context budget fills, so the session never overflows into a lossy compactionWith Paritok, you can cut your token bills 25% on turn 1 to past 85% in long or saturated sessions, and run ~3× more turns in the same context window. Paritok now works with Claude code, Codex, Cursor, and anything OpenAI or Anthropic compatible.Paritok will continue to enhance functionality and adapt user scenarios, while deepening its model capabilities.We’re excited to share this with the PH community and would love your honest feedback.Try Paritok: https://www.paritok.com/Join our Discord to talk with the team: https://discord.gg/SeBJE5EucpThanks for checking us out, and huge thanks to our hunter Chris Messina for hunting us!— Jiayu & Luzhuo