Messaging Limits

LLMs are very expensive, which is why Teta limits the amount of messages that you can send.

​

Free vs Paid Plans comparison

Designed to give users a chance to explore and experience Teta’s features at no cost.

​

Free plan

• You get 5 free daily credits a day
• A maximum of 30 free daily credits a month
• Resets at start of the month

​

Available Paid Plans

Each paid plan includes a set number of allocated messages.

​

How Credit System Works

Teta’s credit system is designed to provide fair pricing based on:

1. Token consumption (input + output + cache)
2. Extended workflows (iterations and multiple AI responses)

​

Token-Based Credits

1 credit = 40,000 tokens processed

How it works:

• For each AI call, we sum: input_tokens + output_tokens + cache_read_tokens
• Credits = total_tokens ÷ 40000 (integer division)
• One credit is charged each time you reach 40k cumulative tokens

Example:

• Call 1: 15,000 tokens → 0 credits
• Call 2: 20,000 tokens → 1 credits (total: 35,000 tokens)
• Call 3: 8,000 tokens → 0 credit (total: 43,000 tokens)
• Call 4: 25,000 tokens → 1 credits (total: 68,000 tokens)
• Call 5: 15,000 tokens → 1 credit (total: 83,000 tokens)

​

Extended Workflow Credits

​

Iteration Credits

Every 5 iterations = +1 credit

• Iterations 1-4: Free (part of base workflow)
• Iterations 5, 10, 15, 20…: +1 credit each

​

Usage Examples

​

Simple Request

User: "Fix this bug in my code"
- 1 user message + 1 AI response
- 8,000 total tokens
- Credits: 1 (under 30k tokens, no extended workflow, + 1 user message)

​

Medium Workflow

User: "Create a new feature with tests"
- 1 user message + 8 AI responses
- 35,000 total tokens
- Credits: 3 (1 for tokens + 1 for response #5 + 1 user message)

​

Complex Workflow

User: "Refactor the entire codebase"
- 1 user message + 25 AI responses
- 20 iterations
- 180,000 total tokens

Credits breakdown:
- User message: 1 credit
- Tokens: 5 credits (180k ÷ 30k = 5 → 5)
- Iterations: 4 credits (iter 5, 10, 15, 20)
- TOTAL: 10 credits

​

Cache-Heavy Workflow

User: "Analyze and optimize the entire project"
- 1 user message + 15 AI responses
- 10 iterations
- Token breakdown:
  * Input: 80,000 tokens
  * Output: 60,000 tokens
  * Cache read: 45,000 tokens
  * TOTAL: 185,000 tokens

Credits breakdown:
- User message: 1 credit
- Tokens: 5 credits (185k ÷ 30k = 5 → 5)
- Iterations: 2 credits (iter 5, 10)
- TOTAL: 8 credits

​

Upgrade Plans

​

FAQ