Can't set up Azure OpenAI as a provider
goose configure
This will update your existing config file
if you prefer, you can edit it directly at /Users/aban/.config/goose/config.yaml
┌ goose-configure
│
◇ What would you like to configure?
│ Configure Providers
│
◇ Which model provider should we use?
│ Azure OpenAI
│
● AZURE_OPENAI_API_KEY is already configured
│
◇ Would you like to update this value?
│ Yes
│
◇ Enter new value for AZURE_OPENAI_API_KEY
│ ▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪
│
● AZURE_OPENAI_ENDPOINT is already configured
│
◇ Would you like to update this value?
│ Yes
│
◇ Enter new value for AZURE_OPENAI_ENDPOINT
│ https://abanagent.openai.azure.com/
│
● AZURE_OPENAI_DEPLOYMENT_NAME is already configured
│
◇ Would you like to update this value?
│ Yes
│
◇ Enter new value for AZURE_OPENAI_DEPLOYMENT_NAME
│ Microsoft.CognitiveServicesOpenAI-20250223113144
│
◇ Enter a model from that provider:
│ gpt-4o
│
◇ Request failed: Request failed with status: 404 Not Found
│
└ Failed to configure provider: init chat completion request with tool did not succeed.
Please provide following information:
- MacOS Sequoia
- Interface: both UI and CLI
- Version: 1.0.7
- Provider & Model:: Azure OpenAI
@thewildofficial can you try a new version. We updated the error hanlding, it should print more details about the 404 error with new version
@thewildofficial let us know if the new version shows more details about the 404 error
Do you know which API version your setup needs? We're using the latest GA release 2024-10-01 (https://learn.microsoft.com/en-us/azure/ai-services/openai/reference)
Hi, same issue, but our company set up an proxy with prefix in base url like https://company.domain/v1/api/ openai/deployments/gpt-4o/chat/completions?api-version=2024-10-21
When this function calls, /v1/api will be overwritten and result in 404
https://github.com/block/goose/blob/fe6cb72677971e33ca7bd8b3823df0658702363b/crates/goose/src/providers/azure.rs#L67-L70
Errors in CLI
◇ Enter new value for AZURE_OPENAI_ENDPOINT
│ https://company.domain/api/v1/
│
● AZURE_OPENAI_DEPLOYMENT_NAME is already configured
│
◇ Would you like to update this value?
│ Yes
│
◇ Enter new value for AZURE_OPENAI_DEPLOYMENT_NAME
│ gpt-4o
│
◇ Enter a model from that provider:
│ gpt-4o
│
◇ Request failed: Unknown error (status 404 Not Found)
│
└ Failed to configure provider: init chat completion request with tool did not succeed.
Errors in UI
Ran into this error: Request failed: Object {"detail": String("Not Found")}.
Please retry if you think this is a transient or recoverable error.
MacOS Sequoia 15.3.1 Interface: both UI and CLI CLI Version: 1.0.10 UI Version: 1.0.9 Provider & Model:: Azure OpenAI
for me the problem was that GPT4o was not enabled on the portal, after doing so, it seemed to work
did you check if it's enabled in azure AI foundry?
for me the problem was that GPT4o was not enabled on the portal, after doing so, it seemed to work
did you check if it's enabled in azure AI foundry?
Yes, other clients (like Roo Cline) with same configuration can work properly.
Settings Screen Capture:
@HeroSizy got it, if you'd like to make a PR with that change so the baseurl you provide doesn't get overwritten, happy to review! otherwise I will try to get to it next week
@alicehau here is the PR, please take your time to review
https://github.com/block/goose/pull/1443
Thanks @HeroSizy , just approved and merged! Going to close out this issue as it seems like we've figured out all the issues in this thread.