#How do I Switch keys in Chat?

15 messages · Page 1 of 1 (latest)

silent crag
#

I have a new Key set up with some money to use for the chat on the website, but I don't see any way to switch over to a different key. When I do anything in chat using the paid models, it just says "max token limit exceeded blah blah blah".

I would assume this is a very simple thing to do... but I can't seem to figure it out.

stuck mortar
#

The chat on the OpenRouter website should deduct your usage automatically from your account total, even if you have no API key generated, those are for external API usage. Or are you talking about a different chat front end?

silent crag
#

huh... that's interesting.

I just got started with all this yesterday, so I probably have no idea what I am doing, but...

I started with Novelcrafter, which is probably the primary application I want to use this for. I set my openrouter account up with my debit card yesterday, although I don't remember seeing anything about a key, but I see two keys in there now... it either happened automatically, or I just completely forgot about it. Anyway, today I wanted to set up a key with a limit on it... but nothing seems to be working.

The first key should be unlimited, which I don't want (but it doesn't seem to work either, if that's the one the openrouter chat on the website uses?), and the second key has $20.

I don't know why nothing is working, I just know that it doesn't seem to be working.

#

Ok, I just verified that it works in Novelcrafter!

But it still doesn't work on the openrouter website itself? 😵‍💫

#

...Okay, no... It uses gpt4o mini, which I believe is payed... but it won't use Claude 3 Opus, which is the main one I am trying to use.

Sorry, I'm trying to troubleshoot.

#

Okay, so no, mini is apparently free? gpt4o pulls up the same error message.

stuck mortar
#

The price for every model is on the model card when you browse the models on the OpenRouter website -> https://openrouter.ai/models
There are a handful free models (rate limited), GPT-4o mini is not one of them. Claude 3 Opus is one of the most expensive models available.

silent crag
#

Um... I think I must have gotten confused somewhere. I thought openrouter served as an alternative to the subscription model for the paid models...

So... If for example, I only used $1.25 worth of Claude 3 opus in a month, I wouldn't be wasting money.

Is that not how it works?

stuck mortar
silent crag
#

GREAT!! That is exactly what I was hoping to find.

I see now that you said GPT-4o mini isn't free? I can use it! That is both exciting and confusing, as I can't use Claude or the regular gpt4o... That makes me think there is a bigger problem going on 😵‍💫

stuck mortar
# silent crag GREAT!! That is exactly what I was hoping to find. I see now that you said GPT-...

https://openrouter.ai/models/openai/gpt-4o-mini will cost you $0.15 / $0.60 (input/output) per million tokens. The price is listed in the providers section (for OpenAI models there is only OpenAI)

GPT-4o mini is OpenAI's newest model after GPT-4 Omni, supporting both text and image inputs with text outputs.

As their most advanced small model, it is many multiples more affordable than other recent frontier models, and more than 60% cheaper than [GPT-3. Run GPT-4o-mini with API

silent crag
#

Mini is working great for me... it does what I tell it to just fine.

But the two more expensive models aren't working at all. Unless they can blow through $20 in one short prompt, I don't think I should be seeing this error?

#

... Oh my god my card is declining and I didn't notice the error message 😒

I didn't successfully pair up my card with the account. That would be the problem, for sure.

stuck mortar
#

I can't help you with billing issues, I am not affiliated with OpenRouter in any way (other than being a paying customer).

silent crag
#

Oh, no, I wouldn't expect you to. I think I got it from here. I really appreciate you taking the time to help, though.