How MiocAI Credits Work: A Transparent Breakdown

We use a credit-based system for some MiocAI features. This system connects usage cost directly to compute demand. Credits might seem abstract at first, so this documentation explains what they represent, how we derived their value, and how you can manage and earn them through weekly, daily, or promotional rewards.

Our purpose is to keep cost visible and correlated with actual processing power.

The Baseline: 1 Credit = 1 Standard Image

The easiest way to understand a credit is through image generation.
When MiocAI renders an image, a powerful GPU (roughly 24GB VRAM per session) is reserved for about 5 to 20 seconds. That compute segment defines our baseline.

One standard image generation costs one credit.
This single metric is the foundation for all other pricing, since every other action consumes more or less compute relative to that baseline.

You can think of a credit as a consistent unit of computational time, proportional to the processing and resource consumption required for a single, standard GPU image generation.

Scaling Up: Upscaling, Video, and Large Models

Each MiocAI task draws on varying levels of processing. Some features are heavier by design. Larger or longer outputs require more GPU time and memory overhead.

Image Upscaling: 3 Credits

Upscaling reconstructs a higher-resolution version of the image instead of merely stretching pixels. The neural models reprocess latent and fine detail data, which triples the GPU task time.

Because each upscale request effectively retrains texture detail on the existing frame, the operation totals about three times the power of a standard generation.

Video Generation: 30 Credits

Video creation multiplies the image generation process over many frames, while maintaining scene and temporal coherence.
The sequential rendering for even short clips uses computational power comparable to 30–60 image generations.
To align cost consistently, this feature consumes 30 credits.

Performance scaling helps preserve predictable pricing between different forms of visual output.

Large Language Models (e.g., Aetherion): 1 Credit

MiocAI includes several model tiers. Standard conversational models are part of your plan with no additional cost.
Advanced models such as Aetherion, which uses around 1.1 trillion parameters, require higher-end computational resources for each message.

Each message to Aetherion averages a cost of 1 credit, though internal complexity can vary with conversation length.

Weekly Credits

Every paid subscription tier provides weekly credit rewards that automatically appear in your account at the renewal checkpoint.
These credits replenish predictable usage without requiring manual action.

Weekly credit awards are time-tracked and applied through your active plan schedule. The cycle resets on renewal, ensuring a steady resource flow for creative work.

Daily Reward Credits

In addition to weekly allocations, you may receive smaller daily bonuses linked to engagement or activity milestones.
These daily credits behave exactly like standard ones and can be used for any generation task.

Their amount may change due to seasonal events, testing programs, or promotions. Daily reward credits encourage consistent activity without requiring purchases.

Admin-Granted Credits and Expiration

Certain cases, such as compensation or promotional gifts, involve manual credit assignment by MiocAI administrators. These credits carry an expiration policy to maintain balance clarity.

Default validity: 1 month from issue date.

After expiry, unused credits are automatically removed from your account.
If an administrator specifies a longer duration at the time of reward, that updated period takes precedence.

Purchased Credit Packs

Purchased credits extend longer than reward-based ones. Credits bought directly through the platform remain valid for 1 year from the purchase date.

Unused balances from those packs expire automatically once that period ends.
This ensures operational stability and prevents indefinite carryover of unused resources.

Credit Usage Order

When you perform a generation action, the system always deducts the credits closest to expiry first.
That means temporary or short-term bonuses are consumed before long-term purchased or subscription credits.

This ordering keeps balances active for the maximum possible duration and prevents wastage.

The Real-World Cost

The cost of purchased credits ranges from $0.05 to $0.10 USD per credit, depending on pack size and region.

Approximate cost breakdown:
* About half of the price reflects GPU and compute cost.
* The remainder supports infrastructure, maintenance, and ongoing development of generation systems.

This transparent structure allows users to understand where each portion of the cost contributes.

Getting the Most from Your Credits

Effective use of credits depends on understanding compute weight across tasks.

• Draft with standard generations. Start creative sessions with lower-cost tests to refine visuals.
• Upscale selectively. Use 3-credit upscaling only for finalized or important renditions.
• Use large models efficiently. Deploy Aetherion for advanced analysis or detailed responses; rely on standard models for lightweight queries.
• Apply video generation strategically. Video outputs have the highest cost. They are best reserved for complex or narrative-heavy applications.

Summary

MiocAI credits represent measurable computational value. Their cost directly corresponds to GPU time and memory resource demand.
Automatic weekly and daily rewards offer recurring access, manual grants provide short-term boosts, and purchased packs ensure longer-term flexibility for active users.
Credits always reduce in order of nearest expiry first, keeping every balance both efficient and transparent.

This structure is designed to maintain fairness, conserve processing load, and make planning your creative work within MiocAI clear and manageable.