no worries have fun I do recommand checking out

tngtech/deepseek-r1t-chimera:free

microsoft/mai-ds-r1:free

nvidia/llama-3.1-nemotron-ultra-253b-v1:free

Thank you for the information about using 'https://openrouter.ai/api' as the URL.
I was struggling with error messages before, but after following this information,

it works perfectly with my API key and deepseek/deepseek-chat-v3-0324:free.
Really appreciate the help!

Seems like you did not put your API key in

I've put a few dollars of credits in an openrouter account, selected "OpenAI/Oobabooga API" as TExt Gen, used "https://openrouter.ai/api" as the URL and "deepseek/deepseek-chat-v3-0324:free" as the key in the text gen option, but every time I start a new game, I get an error. HTTP Error: Unauthorized {"error":{message":"No auth credentials found","code":401}}.. What does that mean? I have zero knowledge of how this stuff works, but I want to use a better text gen than the free default option without subscription fees.

It works!

You were right, I just needed to get rid of ":free"

well you need credits in your account

I'll try it tomorrow, but would it really be possible to permanently run out of credits while using a free model?

"Free Models: Some models are available for free, but even these have strict rate limits (typically 20 requests per minute and 200 requests per day"

try "deepseek/deepseek-chat-v3-0324" if it does not work it means you have no credits in your openrouter account

I typed all of that, but the first request was aborted (I tried to enter a new location)

I tried it one more time and the same thing happened.

The API key was set to V3 and unlimited tokens, so I have no idea why it doesn't work.

No its

https://openrouter.ai/api

KEY

deepseek/deepseek-chat-v3-0324:free

So to sum up:

In URL I should type: https.openrouter/keys (etc) (I can't copy and paste text in the game, so I'll just have to type it maually)

My API key from it below.

And Deepseek V3 at the MODEL?

I mean prompts are just prompts lol just never tell your personal info to ANY ai's and for the api key just reset it

I tried to do it, but it seems only local Ai are possible, and no cloud.

I'm asking because of this:

"Recent cyberattacks have targeted DeepSeek v3, including a critical database leak that exposed over a million records, including system logs, user prompts, and API tokens.

Security researchers have identified flaws in DeepSeek v3's AI models, making them susceptible to jailbreaking techniques. This could potentially allow malicious actors to manipulate the system into generating harmful content or exposing confidential data.

Some data transmissions within the platform have been found to occur without proper encryption, increasing the risk of interception"

Should work but im not sure tho since it may be censored from its training

What about the new free pro model from Gemini?

I now recommand everyone uses the new deepseek/deepseek-chat-v3-0324:free with openrouter

there is an api instruction link on the textgen options menu in game

https://store.steampowered.com/news/app/1889620/view/523076978666898101?l=english

So how do I use minstral ai for example? How do I implement it after subscribing to the mod?

Yeah I apologise its not perfect but yeah its all about finding the right prompt for the right task and well it varies with the ai model you use too. Its quite a pain lol

like "Your left breast presses against the crowbar handle with each heave"!! what the scrungle

ok its actually alright but holy moly does it like talking about the player's boobs

It makes the ai uncensored but ok bro you don't need to announce it

Also generation seems shorter then it was and thats well due to how the dev updated the game I think I cant change the context/output token size