What this error means

openai.NotFoundError: Error code: 404 - Resource not found — Azure OpenAI preview api-version builds incorrect URL with duplicated /openai/deployments/ path is a OpenAI API failure pattern reported for developers trying to fix azure openai sdk build_request adding /deployments/{model} to base_url that already contains /openai/v1/, producing malformed urls like /openai/v1/openai/deployments/gpt-5-chat/chat/completions when using preview api-version. Based on the imported evidence, treat this as a tool-specific troubleshooting page rather than a generic API error.

Why this happens

Specific error pattern discovered while fetching openai-python issue list. Root cause: Azure changed preview API version to use base_url approach, but SDK still appends /deployments/{model} regardless of whether base_url already includes the path prefix. Causes 404 on gpt-5 and other newer models. Requires conditional logic to skip appending when /deployments already present in base_url path.

Common causes

  • Specific error pattern discovered while fetching openai-python issue list. Root cause: Azure changed preview API version to use base_url approach, but SDK still appends /deployments/{model} regardless of whether base_url already includes the path prefix. Causes 404 on gpt-5 and other newer models. Requires conditional logic to skip appending when /deployments already present in base_url path.

Quick fixes

  1. Confirm the exact error signature matches openai.NotFoundError: Error code: 404 - Resource not found — Azure OpenAI preview api-version builds incorrect URL with duplicated /openai/deployments/ path.
  2. Check the OpenAI API account, local tool state, and provider configuration involved in the failing workflow.
  3. Check the build output, project root, and deployment platform configuration before redeploying.

Platform/tool-specific checks

  • Verify the command, editor, extension, or API client that produced the error.
  • Compare local settings with CI, deployment, or editor-level settings when the error appears in only one environment.
  • Avoid deleting credentials, local model data, or project settings until the failing scope is clear.

Step-by-step troubleshooting

  1. Capture the exact error message and the command, editor action, or request that triggered it.
  2. Check whether the failure is account/auth, quota/rate, model/provider, local runtime, or deployment configuration.
  3. Review the source evidence below and compare it with your environment.
  4. Apply one change at a time and rerun the smallest failing action.
  5. Keep the working fix documented for the team or deployment environment.

How to prevent it

  • Keep provider/tool configuration documented.
  • Record non-secret diagnostics such as tool version, provider name, model name, and command path.
  • Add a lightweight check before CI or production workflows depend on the tool.