Json Error adding model

Discussion
1

Hi guys. No idea why I am always getting this error any time I am trying to use any other model after being added. Any time I ve this.
$headers = @{
“Content-Type” = “application/json”
“Authorization” = “Bearer xxxxxxxxxxxxxxxxxxxxxxxxxxx”
}

$body = @{
“messages” = @(
@{
“role” = “system”
“content” = “You are a test assistant.”
},
@{
“role” = “user”
“content” = “Testing. Just say hi and nothing else.”
}
)
“model” = “gpt-4o-mini”
} | ConvertTo-Json

Invoke-WebRequest -Uri “https://api.deepseek.com/beta/chat/completions” -Method Post -Headers $headers -Body $body

In this case I am using deepseek, but it happen as well when I am trying to use qwen2.5-coder:7b running a server ollama on local.

2

Hey, try disabling all models that aren’t related to your API, this should help.

3

Hi. Thank to replied. I did it yesterday though still causing the same issue.

4

Could you share a screenshot of your API settings?

5

deepseek-conf
deepseek-conf820×401 34 KB

Got my api .

6

I have the same settings, and everything is working fine.

image
image1264×506 18.6 KB

7

Did you do anything else before get into Cursos settings? I think I am missing a part. I tried to use qween-code and to this one I running ollama serve and the same thing.
by the way. HAPPY NEW YEAR!!!

9

No, I just entered my API key and changed the base URL. Local models aren’t supported by Cursor.