CEO and co-founder Nick Smirnov says the tool makes repeated build logs retrievable; Hystax measured savings in a benchmark using Codex and GPT-5.5.
By [RuntimeWire Staff](https://runtimewire.com/author/runtimewire-staff)
· Published
Primary source: [PRWeb](https://www.prweb.com/releases/hystax-releases-open-source-token-compressor-that-cuts-ai-coding-agent-costs-by-22-7-without-changing-the-agent-302912355.html)
Why it matters #
Hystax is extending its cloud-cost-management playbook into AI-agent infrastructure: Kompressor is free to inspect and use, while its integration into OptScale AI gives the company a path to sell broader cost and governance tools. Its 22.7% savings claim is company-run and workload-specific, so teams should test it against their own billed traffic.
Hystax released OptScale Kompressor, an MIT-licensed tool that shortens repeated machine output sent by AI coding agents to language models. CEO and co-founder Nick Smirnov says it cut the provider bill by 22.7% in Hystax's benchmark without a statistically significant change in task success.…
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.