What Are Credits?
Credits are ORCFLO's universal currency for workflow execution. Every action in your workflow - from AI model calls to tool usage - consumes credits. This simple, transparent system lets you understand and control your costs while building powerful automations.
Simple Math
1 credit = 100 credit-cents internally. When you see "5 credits" on your dashboard, that's exactly what you have - no hidden decimals or complex calculations.How Credits Work
Monthly Grants
Paid plans receive fresh credits at the start of each billing cycle. Free tier gets a one-time grant.
Per-Step Charging
Credits are deducted when each workflow step completes successfully.
Real-Time Tracking
Monitor usage in your dashboard with detailed transaction history.
Credit Costs
Different operations consume different amounts of credits based on their computational requirements. AI models are grouped into tiers based on capability and cost.
AI Model Costs
AI models charge separately for input tokens (what you send) and output tokens (what the model generates). Costs are shown as input / output credits per 1 million tokens.
Understanding Token Costs
Input tokens are the text you send to the model - your prompt, context, and any data. Output tokens are the text the model generates in response.
Each workflow step has a minimum charge of 1 credit.
Lite Models
Available on all plans
GPT-4o Mini
Budget-friendly GPT-4o for simple tasks
GPT-4.1 Nano
Ultra-affordable GPT-4.1 for simple tasks
GPT-5 Nano
Ultra-affordable GPT-5 for simple tasks
Gemini 2.0 Flash
Previous generation production-ready model
Gemini 2.0 Flash-Lite
Budget-friendly Gemini 2.0 model
Gemini 2.5 Flash-Lite
Fastest and lowest cost Gemini 2.5
Value Models
Available on all plans
Claude Haiku 4.5
Sonnet 4-level coding at 1/3 cost
GPT-4.1 Mini
Fast GPT-4.1 with 1M context
GPT-5 Mini
Fast and affordable GPT-5 variant
o3-mini
Small reasoning model, cost-effective
o4-mini
Fast reasoning for coding and math
Gemini 2.5 Flash
Fast hybrid reasoning with 1M context
Premium Models
Solo, Power plans
Claude Opus 4.5
Most capable Claude for complex reasoning
Claude Sonnet 4.5
Best coding model for complex agents
Claude Sonnet 4
Excellent for complex reasoning
GPT-5.2
Latest GPT-5 with smarter responses
GPT-5.1
GPT-5.1 flagship reasoning model
GPT-4.1
Improved instruction following with 1M context
GPT-4o
OpenAI multimodal model
o3
Advanced reasoning for complex problems
Gemini 2.5 Pro
Google's most capable model for coding
Model Tiers
Free tier includes Lite and Value models. Upgrade to Solo or higher to unlock Premium models.Long Context Pricing
A handful of premium models — such as Gemini 2.5 Pro and Claude Sonnet with 1M context — use tiered pricing. They charge a standard rate for normal workflows and a higher rate when you send them very long prompts or large pasted documents. The higher rate kicks in automatically and applies to the entire run, not just the extra-long portion.
In practice, a long-context run on one of these models can cost roughly twice as muchas a short-context run on the same model. For most everyday workflows this won't come up — but if you're passing in large documents, lengthy conversation histories, or big data payloads, it's worth keeping in mind.
Models with tiered pricing
Standard rate / long-context rate — credits per 1M tokens
Gemini 2.5 Pro
Standard rate for typical prompts; higher rate for long prompts
Gemini 3 Pro
Standard rate for typical prompts; higher rate for long prompts
Claude Sonnet (1M context)
Standard rate for typical prompts; higher rate for long prompts
The whole run is billed at the higher rate
When you send a very long prompt to one of these models, bothyour input and the model's response are charged at the higher rate — not just the extra-long portion. For most workflows this won't matter, but long documents or large conversation histories can roughly double the credit cost.The model selector shows two prices side-by-side (e.g. 125 / 250 on long context) for these models. If you typically work with shorter inputs you'll pay the standard rate; long conversations or large pasted documents will cross into the higher rate. Flat-priced models show a single rate with no long-context surcharge.
Tool Costs
Credits per tool invocation. MCP Tools available on all plans.
Web Tools
credits / callTavily Search
Search the web using Tavily API
Tavily Crawl
Deep crawl websites
Tavily Extract
Extract structured data from pages
HTTP Request
Make HTTP requests to APIs
Communication
credits / callSend Email
Send emails with optional formatting
Development
credits / callPython
Write and run Python code
Math
Mathematical calculations
Credit Grants
Each plan includes a credit allocation. Paid plans receive monthly grants that refresh at the start of each billing cycle. The free tier receives a one-time grant at sign-up.
| Plan | Credits | Price | Overage |
|---|---|---|---|
| Free | 500 (one-time) | $0 | Not available |
| Solo | 1,500/month | $15/month | $10/1K credits |
| Power | 3,600/month | $30/month | $8.33/1K credits |
Overage Billing
Solo and Power plans can enable overage usage when you exceed your monthly allocation. Overage is charged nightly at your plan's rate. Free tier workflows pause until you upgrade.Tracking Your Usage
Stay on top of your credit consumption with these tools:
- Credit Balance: View your remaining credits, usage percentage, and days until refresh from your dashboard.
- Transaction History: See every credit charge and grant with timestamps in your billing settings.
- Low Credit Warnings:Get notified when you've used 75% or more of your credits.
Best Practices
- Choose the Right Model Tier: Use Lite models for simple tasks. Save Premium models for complex reasoning.
- Test with Free Credits: Your 500 free credits never expire. Use them to prototype before upgrading.
- Monitor Weekly: Check your transaction history to understand usage patterns and optimize.
- Upgrade for Better Rates: Higher plans get better overage rates: Power at $8.33/1K vs Solo at $10/1K.