problem with integrating OpenAi to my Trados 2024 project

Hello!

I have installed Trados 2024 and am encountering an issue when trying to integrate OpenAI into my project. I have a subscription to OpenAI, which I regularly use with GPT-4o, and I have created a valid API key.

However, when attempting to add OpenAI to my project (by clicking on "EDIT"), Trados only offers a connection to GPT-3.5 Turbo. After entering my API key, I receive the following error message: "CONNECTION FAILED", with error code: -2146233088.

Additionally, if I try to create a new connection model (GPT-4o) and click "OK," nothing happens. When I test the connection, I encounter the same error code.

Is this issue occurring because my API key is for GPT-4o, while Trados only supports GPT-3.5 Turbo? I would appreciate clarification on whether Trados 2024 supports GPT-4o and guidance on how to resolve this issue.

Thank you in advance for your assistance.

add
verb [with obj]
1) join (something) to something else so as to increase the size, number, or amount
[no obj.] (add up) increase in amount, number, or degree
2) put in (an additional element, ingredient, etc.)
contribute (an enhancing quality) to something
3) put together (two or more numbers or amounts) to calculate their total value
[no obj.] (add up to) amount to
[no obj.] [usu. with negative] (add up) informal seem reasonable or consistent; make sense
4) [reporting verb] say as a further remark: [with direct speech]
Open in ABBYY Lingvo European Dictionary
emoji
Parents Reply
  •  

    I have a paid subscription to ChatGPT, for which I pay 20 USD per month, and I use GPT-4o. Are you saying that I need a different account?

    Yes.  ChatGPT does not entitle you to work with the API.  An account that gives access to ChatGPT and an API are two separate services with different purposes and pricing models.

    • ChatGPT access: This allows you to interact with the model directly through a user interface. It's a standalone service, with its own subscription and pretty much gives you unlimited use for a flat fee.

    • API access: This is designed for developers who want to integrate GPT into their own applications or systems. It involves sending requests from external software to the GPT model and getting responses programmatically.  If you want to use such an integration you will need to provide an appropriate API key so you can billed for the number of times you use it.

    These services are priced separately because:

    1. Different usage types: ChatGPT access is personal and direct, while the API is for building external apps and handling more automated, large-scale interactions.
    2. Infrastructure costs: Providing API access requires different infrastructure and resources, which scale with usage in ways that personal interaction through the ChatGPT UI doesn't.

    In short, even if you're paying for ChatGPT access, the API service is considered a different product by OpenAI with separate costs due to its distinct usage and technical demands.

    Maybe this page will be helpful for you?

    https://help.openai.com/en/articles/9038407-how-can-i-set-up-billing-for-my-account

    Paul Filkin | RWS Group

    ________________________
    Design your own training!

    You've done the courses and still need to go a little further, or still not clear? 
    Tell us what you need in our Community Solutions Hub

    emoji
Children