Caveman
why use many token when few do trick
Product HuntOpen SourceLaunched Aug 13, 2026166#8 of the day
About Caveman
One command wraps Claude Code, Codex, Hermes, and more with a local proxy that compresses logs, tool output, and files before every provider call. In a pinned 54-run benchmark: 33.2% fewer input tokens with 18/18 correctness checks. Caveman can also run any existing agent skill with ~70% fewer tokens by loading text as images. Built on an open-source ecosystem with 97K+ GitHub stars.
The person behind Caveman
More Open Source launches
InyonAnswer your coding agents from your phoneOct 23, 2026 · 1 person
NanotypeA private online text editor with markdown. Open and type.Oct 23, 2026 · 1 person- Whiteboardthe open-source IDE for thoughtful software designOct 22, 2026 · 1 person
Tetriz Engg Intelligence + ConnectorsConnect coding sessions to PRs, issues, quality, and ROIOct 22, 2026 · 1 person
TessaryStop your agents from failing silently in productionOct 22, 2026 · 1 person
LLM Test ServerYour LLM tests, without the providerOct 22, 2026 · 1 person
Made Caveman? Ask us to remove this listing or your details.