Models

Chat Models & Limits

API Models & Pricing

How Credits Are Calculated

Credits are charged per message in your conversation array. This includes both user messages and assistant responses.

Example: A conversation with 4 messages using GPT OSS 20B:

[
  { "role": "user", "content": "How are you?" },
  { "role": "assistant", "content": "I'm well, thank you!" },
  { "role": "user", "content": "What's the weather like?" },
  { "role": "assistant", "content": "I don't have access to weather data." }
]

Cost: 4 messages × 0.1 credits = 0.4 credits

Reducing Costs

• Summarize long chats: For conversations over 10 messages, summarize the history and start fresh with the summary as context.
• Use cheaper models for simple tasks: GPT OSS 20B (0.1 credits) handles most tasks well. Save premium models for complex reasoning.
• Cache frequent responses: Store and reuse responses for common queries instead of making repeated API calls.
• Set lower maxTokens: Limit response length when you only need short answers.
• Skip unnecessary history: Only include previous messages when the model actually needs context.

Tools