Densely packs logs, JSON, tool outputs and code into 2×–8× fewer tokens — with guaranteed byte-exact reconstruction, verified by sha256 on every decompress. Nothing is summarized. Nothing expires. Nothing is ever dropped.
Densely reclaims the difference. Compressed bytes are re-encoded as a stream of 65,536 single-token English words — the o200k pre-tokenizer never merges across word boundaries, so every carrier word costs exactly one token and carries two bytes of compressed data. That's 16 of the ~17.6 available bits: 91% of channel capacity.
A hook compresses every large tool output on the fly — no workflow changes. A savings ledger shows tokens saved today and this week, so the effect on your usage cap is visible, not vibes.
Works in Claude Code and Cursor today: compress_file, search, expand. The agent searches inside compressed payloads and expands exact slices — without inflating the whole thing back into context.
pip install densely, three commands. Compressed payloads are plain text: they survive context compaction, session export, and machine moves. No cache, no TTL, nothing to expire.
Qwen2.5-Coder + arithmetic coding pushes code from 2.02× to 7.27× — 86.3% fewer tokens, still byte-exact. A record for lossless compression on code.
Tools like Headroom keep compressed content readable by the model — by selecting what to keep. Densely keeps every byte — but the payload is opaque until expanded. Different trade, different jobs.
Open source, MIT, runs locally. Benchmarks ship with the repo.
A hosted tier for teams, in development — it ships to this list first:
Joining now gets you early access in signup order, founding-user pricing locked in, and a direct line to us while we build. The MIT library stays free either way — Pro is hosted convenience, nothing gets gated.