GPT-Realtime-2

lancerichards

Platinum Partner
Advanced Certified
Joined
Aug 14, 2024
Messages
159
Reaction score
195
Hi Team

Is GPT-Realtime-2 going to be available in the final U9 release?

Best,
Lance
 
Hi Lance,

Not in U9 - there are known issues with Realtime 2 reported in the OpenAI community. We'll stick with Realtime 1.5 for now and revisit once those are resolved.

Best,
Michael
 
any news on that?
 
Are there any plans to allow newer AI models to be added without waiting for future Updates? For example, newer AI models sometimes come out quite fast and GPT-Realtime-2, GPT-Realtime-2.1, and GPT-Realtime-2.1 mini have all released since this thread was first made, but we are still on GPT-Realtime-1.5 in Update 9
 
Are there any plans to allow newer AI models to be added without waiting for future Updates? For example, newer AI models sometimes come out quite fast and GPT-Realtime-2, GPT-Realtime-2.1, and GPT-Realtime-2.1 mini have all released since this thread was first made, but we are still on GPT-Realtime-1.5 in Update 9
In U10, the latest models will be available.
 
Sounds much better. :)
Well done team.
 
In U10, the latest models will be available.
GPT-Realtime-2 is in the Update 10 Alpha, but GPT-Realtime-2.1, and GPT-Realtime-2.1 mini are not. Any chance they will be added?
 
They will be in the beta coming soon.
 
Sorry to kind of hijack this topic, but I am having issues with both these models at the moment. And this discussion was about different models, so it seemed perhaps related.
I'm using the new U10 alpha, but the issues might have been there earlier. We're still not live with any AI agent, it's all experimental at the moment for us.

But the problem is, both real-time models don't seem to exist.
The 3CX log notes:
-The model `gpt-realtime-1.5` does not exist or you do not have access to it.
-The model `gpt-realtime-2` does not exist or you do not have access to it.

When I use PS to lookup the available models, I get this:

PS C:\> (Invoke-RestMethod `
>> -Uri "https://api.openai.com/v1/models" `
>> -Headers @{ Authorization = "Bearer $env:OPENAI_API_KEY" }
>> ).data |
>> Where-Object { $_.id -like "*realtime*" } |
>> Select-Object id

id
--
gpt-realtime
gpt-realtime-mini


PS C:\> Invoke-RestMethod `
>> -Uri "https://api.openai.com/v1/models/gpt-realtime" `
>> -Headers @{ Authorization = "Bearer $env:OPENAI_API_KEY" }

id object created owned_by
-- ------ ------- --------
gpt-realtime model 1756271701 system


PS C:\> Invoke-RestMethod `
>> -Uri "https://api.openai.com/v1/models/gpt-realtime-1.5" `
>> -Headers @{ Authorization = "Bearer $env:OPENAI_API_KEY" }
Invoke-RestMethod : {
"error": {
"message": "That model does not exist",
"type": "invalid_request_error",
"param": "id",
"code": null
}
}





Should 3CX just leave out the version numbers in the model selection or am I doing something wrong now?
I has worked earlier, but the last time I have been working on this was a few months ago.
 
  • Like
Reactions: nikolascx
Sorry to kind of hijack this topic, but I am having issues with both these models at the moment. And this discussion was about different models, so it seemed perhaps related.
I'm using the new U10 alpha, but the issues might have been there earlier. We're still not live with any AI agent, it's all experimental at the moment for us.

But the problem is, both real-time models don't seem to exist.
The 3CX log notes:
-The model `gpt-realtime-1.5` does not exist or you do not have access to it.
-The model `gpt-realtime-2` does not exist or you do not have access to it.

When I use PS to lookup the available models, I get this:

PS C:\> (Invoke-RestMethod `
>> -Uri "https://api.openai.com/v1/models" `
>> -Headers @{ Authorization = "Bearer $env:OPENAI_API_KEY" }
>> ).data |
>> Where-Object { $_.id -like "*realtime*" } |
>> Select-Object id

id
--
gpt-realtime
gpt-realtime-mini


PS C:\> Invoke-RestMethod `
>> -Uri "https://api.openai.com/v1/models/gpt-realtime" `
>> -Headers @{ Authorization = "Bearer $env:OPENAI_API_KEY" }

id object created owned_by
-- ------ ------- --------
gpt-realtime model 1756271701 system


PS C:\> Invoke-RestMethod `
>> -Uri "https://api.openai.com/v1/models/gpt-realtime-1.5" `
>> -Headers @{ Authorization = "Bearer $env:OPENAI_API_KEY" }
Invoke-RestMethod : {
"error": {
"message": "That model does not exist",
"type": "invalid_request_error",
"param": "id",
"code": null
}
}


Should 3CX just leave out the version numbers in the model selection or am I doing something wrong now?
I has worked earlier, but the last time I have been working on this was a few months ago.

The problem is that your openai Key does not have access to use these models.
It has access to gpt-realtime and gpt-realtime-mini but now we removed those models. (there is no point to keep them - they are old and buggy)

You need to go in your openai API and allow all models.

Secure the API Key and allow all models. Because AI is moving fast and we cannot keep up with all these changes.

Start from here https://platform.openai.com/settings/organization/limits and check from this point. Are you by any chance on the free tier?
 
  • Like
Reactions: Evolute IT

Forum statistics

Threads
111,823
Messages
589,219
Members
164,650
Latest member
Kellenwalsh