Make the Most of Your Magicdoor Usage Balance
Last updated: October 7, 2026.
Your usage balance is not a fixed number of answers. Actual charges depend on tokens, context, tools and the selected model. Check current model rates before choosing a workflow.
Start with a representative prompt
GPT-6 Luna costs $0.10 per million input tokens and $0.50 per million output tokens at standard rates. Try a short task, verify the answer and review its actual usage. On a harder task, compare GPT-6.1 Sol or Claude Sonnet 5.5 rather than assuming a higher rate always improves the result.
If the question needs current sources, choose a search-enabled chat model such as Claude Haiku 5.5 and open the cited original pages. The old Perplexity Reasoning and Deep Research (pplx) picker options are retired.
Know what your plan includes
Basic costs $6/month with $2 included usage balance in its first month and $1 per renewal. Plus costs $12/month with $7 monthly usage balance. Additional usage is billed separately; purchased top-ups are separate from monthly included balance. Check the pricing page for the current terms. Avoid treating historical per-conversation examples as a promise of how long a balance will last.
Try current models on Magicdoor
FAQ
Which lower-rate chat model can I try first?
GPT-6 Luna has standard rates of $0.10 input and $0.50 output per million tokens. Try a representative prompt and check your usage rather than assuming a fixed number of conversations.
How much included balance comes with Basic?
Basic costs $6/month and includes $2 usage balance in the first month, then $1 at each renewal. Plus costs $12/month and includes $7 monthly usage balance.
What if I need current web sources?
Select a current search-enabled model such as Claude Haiku 5.5 or GPT-6.1 Sol and verify cited sources. The retired Perplexity picker choices are not separate current models.
Related Resources
Gemini Flash Guide: Historical Magicdoor Availability
Gemini Flash is retiring from the Magicdoor chat picker. Learn which current low-cost models to compare and how existing conversations are handled.
Claude Haiku 5.5 Guide on magicdoor.ai
Compare Claude Haiku 5.5 with GPT-6 Luna and Sonnet 5.5 using token prices, long-context limits, and Anthropic's published evaluations.
Gemini 3.1 Pro Historical Guide
Gemini 3.1 Pro is no longer selectable on magicdoor.ai. See its former pricing and current model alternatives without conflating Google product access with Magicdoor.
Image Model Comparison - When to Use Each magicdoor.ai Image Model
Practical guide to choosing between magicdoor.ai's current image models for generation, editing, background cleanup, higher-resolution output, and upscaling.