Pricing

Simple,
transparent plans.

Free tier, Pro after a 30-day trial, Premium for teams — and the questions people ask most.

← Back to the overview

Pricing

Simple, transparent plans

Monthly plans include a 30-day free trial — $0 today. Prefer flexibility? Weekly and quarterly billing available.

Pro
$4.99/mo
or $1.99/week · $12/quarter (~$4/mo — save 20%)

For developers running agent sessions daily. Unlimited prompts, multi-session monitoring.

30-day free trial — cancel anytime
  • Unlimited optimizations
  • 3 connected sessions
  • 2 devices
  • All 3 optimization modes
  • Agent monitoring + duplicate detection
  • Auto-replace & Send-mode
  • CLAUDE.md rule generation
Premium
$99/mo

For teams and power users. Unlimited everything, priority support.

30-day free trial — cancel anytime
  • Unlimited optimizations
  • Unlimited connected sessions
  • Unlimited devices
  • All 3 optimization modes
  • Full agent analytics + rule generation
  • Auto-replace & Send-mode
  • Priority support
FAQ

Frequently asked questions

Everything you need to know about token optimization and how Terse saves you money.

What is token optimization?
Token optimization is the process of reducing the number of tokens in AI prompts and outputs without losing meaning. Terse uses 35+ techniques — including spell correction, filler removal, prompt compression, and semantic deduplication — to cut token usage by 40-70%, directly lowering AI API costs. Read the full guide →
How much can Terse save on AI costs?
Terse reduces token usage by 40-70% on verbose prompts and up to 89% on CLI output noise. A typical 2-hour coding session generates ~210K tokens of raw output, which Terse compresses to ~23K. Combined with duplicate detection and redundant read flagging, total session costs drop 3-5x. See the cost breakdown →
Which AI tools does Terse work with?
Terse auto-detects Claude Code, Cursor, OpenClaw, Aider, and any terminal-based AI agent via process scanning. It also works with browser-based tools like ChatGPT, Claude.ai, and Gemini through macOS Accessibility API integration with Chrome and Safari. See agent monitoring →
How does prompt compression work?
Terse runs a 7-stage pipeline: spell correction (400+ typo fixes), whitespace normalization, pattern optimization (130+ rules), redundancy elimination, NLP analysis, telegraph compression, and final cleanup. Code blocks, URLs, and technical terms are protected throughout.
Does optimization reduce AI output quality?
No. Token optimization removes noise — filler words, hedging, redundant phrases, typos — without changing intent. Research from LLMLingua (EMNLP 2023) shows compressed prompts maintain or improve output quality because models receive cleaner, more focused instructions.
What’s the difference from prompt engineering?
Prompt engineering focuses on crafting better instructions for better AI outputs. Token optimization reduces the cost of those instructions by removing waste — filler, typos, redundancy — without changing meaning. Terse handles optimization automatically so you can focus on engineering for quality. Learn more →
Is Terse free to use?
Yes. Both plans include a 30-day free trial — no charge until your trial ends. Pro ($4.99/mo) includes unlimited optimizations, 3 sessions, agent monitoring, and CLAUDE.md generation. Premium ($99/mo) includes unlimited sessions, devices, and priority support. Cancel anytime.
How do AI tokens affect cost?
AI models charge per token (~4 characters each). Claude Opus costs $15-$75 per million tokens, GPT-4o costs $2.50-$10. A single agent session can consume 200K+ tokens, costing $3-$15. Heavy users spend $200-500+/month on API costs alone. See the full pricing comparison →
What is the Token Exchange?
The Token Exchange is a marketplace where users trade unused AI API tokens. Sellers list their API keys at a discount, buyers get cheaper access. Terse runs a proxy that optimizes every request before forwarding, so the actual API cost is 30-60% less. Terse takes a 15% commission. Your API keys are encrypted with AES-256 and never exposed to buyers.
How do I buy or sell tokens?
Sign in at terseai.org/marketplace. To sell: paste your API key, drag a slider to set your discount, done. To buy: top up your balance, generate a Terse API key, and use it in any SDK — just set the base URL to Terse's proxy. One terminal command to configure.

Stop wasting
tokens and money.

Optimize every prompt. Monitor every agent session. And give your whole team visibility with Terse Cloud — analytics by developer, project, and tool.

macOS Windows VS Code Chrome Terse Cloud Join our Slack

First Launch — macOS Security Step (one-time only)

macOS blocks unsigned apps by default. Pick whichever method works for you:

A System Settings — easiest, no Terminal needed
1. Open Terse — click OK on the security warning
2. Open System Settings → Privacy & Security
3. Scroll to the Security section — click "Open Anyway" next to Terse
4. Confirm in the popup — done. Terse opens normally from now on.
B Right-click to open — works on macOS Ventura and earlier
1. In Finder, right-click (or Control+click) on Terse.app
2. Choose Open from the menu
3. Click Open in the dialog that appears
Note: macOS Sequoia (15+) removed this option — use Method A instead.
C Terminal command — one command, works on all versions
Drag Terse to /Applications, then paste this in Terminal:
xattr -cr /Applications/Terse.app && /Applications/Terse.app/Contents/MacOS/terse 2>/dev/null &
100% on-device Zero latency