2-3 Prompts is $10?? Something isn't right

[brisk ravine]

I'm an AI assistant - my answers may be incorrect and I cannot see screenshots. Please verify information with the OpenRouter docs.

To help get to the bottom of your billing issue, I need a bit more information:

• What exact model(s) are you using for these prompts? (e.g., openai/gpt-4o-2024-08-06)
• Are you sending long prompts or using image input?
• Can you check your Activity page and share the per-prompt credit usage or any unusual charges you see?

These details will help diagnose if the charges are correct based on model pricing, or if there's a billing error that needs escalation. For account or billing issues, please email [email protected], as billing is not handled on Discord.

For reference: "OpenRouter uses a credit system denominated in USD. You can top up manually or enable auto top up... Use the Activity page or the credits API for balance and remaining credits" Support - OpenRouter.

-# This is an automated response powered by Answer HQ. Use !mute to stop the bot from responding.

[little minnow]

2-3 Prompts is $10?? Something isn't right

[brisk ravine]

Based on your description, it sounds like you're using high-end models such as openai/gpt-4-turbo-preview, openai/gpt-4o-audio-preview, or similar. These models can be very expensive if your prompts (inputs and outputs) are large.

For example:

• openai/gpt-4-turbo-preview is priced at "$10 per million input tokens, $30 per million output tokens" GPT-4 Turbo Pricing
• openai/gpt-4o-audio-preview is "$2.50/M input tokens, $10/M output tokens, $40/M audio tokens" GPT-4o Audio Pricing

If your prompts or outputs are especially lengthy, a single prompt can use thousands of tokens, which adds up quickly. For example, 100,000 output tokens at $10/M would cost $1, and 300,000 output tokens would be $3. If each conversation is very long or includes images/audio, the cost goes up even faster.

To see exactly what you're being charged per prompt, check your Activity page. If you believe there is a billing error, you need to email [email protected], as billing issues are not handled on Discord.

For more on model pricing and how charges are calculated:

• "Input $10 / M tokens, Output $30 / M tokens" GPT-4 Turbo Pricing
• "Input $2.50 / M tokens, Output $10 / M tokens, Audio tokens $40 / M" GPT-4o Audio Pricing

Let me know if you have your model name and example prompt sizes, and I can help you estimate expected costs.

-# This is an automated response powered by Answer HQ. Use !mute to stop the bot from responding.

[little minnow]

I need help. Im using 2-3 prompts and having to re-up -- [email protected]

[gritty flint]

what model are you using?

[little minnow]

anthropic/claude-sonnet-4:1m

[gritty flint]

can you show me a screenshot of your activity page?

[gritty flint]

i was refering to the other activity page

which shows model requests

[gritty flint]

the cost looks about right

[wet carbon]

above 200k tokens, the price goes up

also, you might not have caching enabled on Cline

sorry, it does actually.

thats about right

[little minnow]

so if the convo is getting to big, it exponentially gets more expensive?

[wet carbon]

yes

[little minnow]

ahh

[wet carbon]

that, and also because past 200k tokens, the cost per token increases FOR this specific model

[little minnow]

thank you