[✅ Solved] Credit use

[snow moss]

I adjusted my max tokens in the app I use and it still uses 8 credits per request

[vapid pagoda]

Thanks for sharing the screenshot! The number of credits used per request is based on how many tokens (words/characters) you send in plus how many tokens the model generates as output, not just your "max tokens" setting in the app. Even if you lower the max tokens, if your input plus expected output is large, the credit usage may still be higher than 1 credit.

For DeepSeek-V3.2, the credit use is proportional to the total number of tokens (input + output). Each model has its own ratio for credits-per-token; you can check those details on the Models page.

What you can do:

• Try sending shorter messages and set lower "Max Tokens" to reduce output.
• Each app sometimes sends extra prompts or history with every request, which also increases credit use.
• Check the actual token count in your app, if possible, and compare that to what’s shown in your request logs on the website.

For more details about how credits are calculated, visit: https://pawan.krd/docs/usage#credits

If you want to optimize further, see full details and recommended settings for each model on the Models page.

If you need to track your credit usage more closely, you can always check your usage logs on the website.

Let me know if you need step-by-step help with settings!

Please note: This is an automated response from @vapid pagoda. While it strive to provide accurate assistance, It may occasionally make mistakes. If you find any inaccuracies or need further clarification, feel free to wait for our community helpers to give you further guidance. Thank you.

[snow moss]

Put my settings to the recommended, still using 8 credits per request even though tokens are really low

[vapid pagoda]

[✅ Solved] Credit use

✅ Thread has been marked as resolved by @snow moss!
If you still need help, please create a new thread in #1122465350110412841.