Henji vs Paritok: Features, Pricing & Which Is Better (2026)

A side-by-side comparison of Henji and Paritok — features, pricing, and ideal use cases — to help you decide which AI tool fits your workflow.

Henji

Freemium

Mac app that drafts chat and email replies in your own voice across Slack, LINE, Gmail, and Messages.

Voice Matching: Learns your usual tone and phrasing over time so replies read as you-ish rather than AI-ish.
Tone Modes: Switch between Polite, Casual, Team, and Friends styles so each reply fits the relationship and channel.
Multi-Channel Coverage: Works across Slack, LINE, Gmail, and Messages so chat and email replies are handled in one place.
Scribble-to-Reply: Type a short note or intent and Henji expands it into a complete, context-aware message.
Multilingual: Supports multiple languages including English and Japanese for replies.

Faster Messaging: Knocking out quick chat and email replies during a busy day without sounding robotic.
Difficult Replies: Politely declining requests or negotiating deadlines while keeping the tone warm.
Team Communication: Keeping internal Slack threads fast and to the point with a team-appropriate tone.
Cross-Language Correspondence: Drafting replies in English or Japanese for international contacts.

Freemium

Non-destructive compression gateway that drops between coding agents and LLMs to cut input tokens by up to 85% without changing the agent.

Drop-In Gateway: One environment variable (ANTHROPIC_BASE_URL) reroutes your agent through Paritok — no agent, prompt, or tool changes.
Tool Schema Compression: 46-schema tool blocks (~29K tokens) drop to ~8K per turn by keeping relevant tools and stubbing the rest, frozen per conversation for cache stability.
Code-Native 4B Model: A 4B compression model trained on 45K real agent trajectories keeps identifiers, paths, and errors while shrinking file reads and outputs to ~26% of original.
read_original Recall: Every compressed segment is tagged; the agent asks read_original(ref) and gets the exact bytes locally without spending an extra turn.
Stale History Summarization: Turns beyond a configurable recent window get summarized once when your context budget fills, so recent turns stay pristine and overflows drop to zero.
Multi-Agent Compatibility: Works today with Claude Code, Cursor, Codex, OpenHands, and any OpenAI-compatible upstream — Anthropic and OpenAI both supported.
Compounding Savings: Saved share grows across a session — 25% at 1 turn, 54% at 10 turns, 63% at 20 turns — against a 96,500-token baseline.
Open Weights and Benchmark: SWE-bench Lite floor of 86.5% quality retained at 25.7% compression rate, with weights and training pipeline published.

MCP-Heavy Workflows: Cut input bills for agents that ship 70+ MCP tool schemas on every turn.
Long Coding Sessions: Run 3× longer coding-agent sessions before context saturation forces a hard compact.
Bill Reduction: Estimate 54% off input tokens on a 5-developer team at 20-turn Claude Sonnet sessions (~$6,550/year).
Self-Hosted Privacy: Route agent traffic through your own hardware with no data leaving your network — 8GB GPU is enough.
Enterprise Cost Governance: Add compression at the gateway layer so all coding agents on the team benefit without engineering per-agent.
Cursor/Codex/Claude Code Fleet: Standardize compression across a mixed toolchain of coding agents behind one gateway.