Assistant CoreAssistant Core
Billing

Credits

Understand the credit system — how credits are calculated, consumed, and tracked.

What Are Credits?

Credits are the unified currency on Assistant Core. Every AI operation — chat, voice, realtime — consumes credits from your wallet. More expensive models cost more credits, cheaper models cost fewer — you only pay for what you actually use.

Benefits of Credits

  • Pay for what you use — Premium models cost more credits, efficient models cost fewer
  • One wallet for everything — Chat, voice, and realtime all share the same credits
  • Flexible monthly budget — Use your credits however you want throughout the month

How Are Credits Calculated?

Each AI operation is charged credits based on the model and usage:

Chat (LLM)

Credits are calculated by token count (input and output):

FactorDescription
Input tokensYour message + conversation context
Output tokensThe AI's response
Cached tokensTokens from previously processed context

Example: A typical chat message (~500 input + 300 output tokens) with GPT-4.1 Mini costs about 1–2 credits.

Voice

TypeHow It's Charged
Speech to Text (ASR)By audio duration (minutes or hours)
Text to Speech (TTS)By character count or audio duration
Realtime VoiceBy session duration (minutes)

Utility & Embedding

Some background features also consume credits in very small amounts:

FeatureWhen
Follow-up suggestionsAfter each chat turn (if enabled)
Title generationFirst message of each conversation
Long MemoryExtracting key information from conversations
Knowledge Base searchFinding relevant content in your documents
Memory recallRecalling previously saved information

Utility and embedding costs are very small (typically under 1 credit per operation) and are calculated automatically — you don't need to worry about them.

Image Generation

Credits are charged by image count and quality:

ModelQualityCredits / image (estimate)
GPT Image MiniAuto / Low~8
GPT Image MiniMedium~23
GPT Image MiniHigh~54
Gemini Flash ImageDefault~59

Image generation is more expensive than regular chat. Each image costs roughly 8–60 credits depending on model and quality.

Credits are charged by number of queries:

TypeCredits / query
Search (web, map, video, scholar)~2
Webpage scrape~3

Estimated Credits Per Model

The number of credits per chat message depends on the model you choose and conversation length. Here are estimates:

ModelCredits / message (estimate)
GPT-4.1 Mini~2–5
Gemini 2.5 Flash~2–5
DeepSeek V4 Flash~2–5
GPT-4.1~5–15
Claude Sonnet 4.5~10–25
GPT-5.4~15–40

These are estimates. Longer conversations build up more context, which increases credit cost per message. Tool calls also add to the total.

Tracking Your Credits

Check Your Balance

  1. Go to My AccountMy Subscription
  2. The Credits bar shows:
    • Used — Credits consumed in the current period
    • Remaining — Available credits
    • Limit — Total credits granted for this period

When Credits Run Out

When credits reach zero, AI operations are blocked immediately:

  • Chat — Cannot send new messages
  • Voice — Cannot start new voice sessions
  • Realtime — Realtime voice sessions are rejected

Zero credits = AI usage paused. Upgrade your plan for more credits, or wait for the next billing period to receive a fresh allocation.

Credit Lifecycle

Start of billing period
  └─ Credits granted per plan (no rollover from previous period)

During the period
  └─ Each AI operation deducts credits from wallet

End of period
  └─ Remaining credits expire → reset to 0
  └─ Fresh credits granted for the next period

Credits do not roll over between billing periods. At the start of each period, your wallet is reset and refilled with exactly the credits your plan provides.

Need More Credits?

Compare plans to find the right fit, or contact us for an Enterprise plan with unlimited credits.

On this page