Understanding AI Costs
How much does it cost to use AI in QuickContract?
How AI pricing works
Cloud-based AI providers charge based on tokens — the units of text that models process. Roughly speaking, one token equals about 4 characters of English text. A typical contract contains 2,000–5,000 tokens of output, plus the input tokens from your prompts, meeting transcripts, and business profile data.
You pay for both input tokens (what you send to the model) and output tokens (what the model generates). Output tokens are typically 2–5x more expensive than input tokens. QuickContract does not add any markup — you pay only what your AI provider charges.
Cost per contract
A typical contract generation costs between $0.02 and $0.10 depending on the model and contract length. Here is a rough breakdown by activity:
| Activity | Estimated cost |
|---|---|
| Generate a standard contract | $0.02 – $0.10 |
| QuickEdit a section | $0.01 – $0.03 |
| Legal Advisor chat message | $0.005 – $0.02 |
| Contract comparison | $0.03 – $0.08 |
| AI agent task | $0.05 – $0.15 |
| Generate clause variant | $0.01 – $0.03 |
These estimates assume mid-tier models like Claude Sonnet or GPT-4o. Premium models (Claude Opus, GPT-4) cost more. Budget models (Llama, Gemini Flash) cost less.
$5 on OpenRouter typically lasts weeks of normal use. That is enough to generate 50–200 contracts, run dozens of QuickEdit operations, and have many Legal Advisor conversations. Most users top up once and do not think about costs for a long time.
Tracking your spending
When OpenRouter is your active provider, QuickContract displays your remaining balance in the sidebar. This updates automatically after each AI operation so you always know where you stand.
For Anthropic, OpenAI, and Gemini, check your usage on each provider's dashboard. These providers bill to your payment method at the end of each billing cycle or when you hit a spending threshold.
Reducing costs
- Use a mid-tier model — Claude Sonnet and GPT-4o offer excellent quality at a fraction of the cost of premium models
- Use templates — generating from a template uses fewer tokens because the AI has a clear structure to follow
- Use Ollama for drafts — run a free local model for initial drafts, then switch to a cloud model for final polishing
- Be specific in prompts — clear, detailed input leads to better output on the first try, avoiding costly regeneration
Free option: Ollama
If AI costs are a concern, Ollama lets you run AI completely free on your own hardware. There are no per-token charges and no accounts to manage. The trade-off is slower generation speed and somewhat lower quality compared to the best cloud models.