Gemini API: `404 models/gemini-1.5-flash is not found for API version v1beta, or is not supported for generateContent. Call ListModels to see the list of available models`

发生场景

Source(s): https://stackoverflow.com/questions/79779187 , https://stackoverflow.com/questions/79477167 Signal: 45,853 + 8,807 views. Top answer 42 votes; the second question has an accepted answer; a comment on another answer says "this solution worked for me". Status: reported working by source, not reproduced by us.

问题

Code that used to work (genai.GenerativeModel("gemini-1.5-flash"), or gemini-pro in the Node SDK) starts returning 404 for every call. The same happens for old embedding names such as text-embedding-004.

报错原文（已脱敏/规范化）

404 models/gemini-1.5-flash is not found for API version v1beta, or is not supported for generateContent. Call ListModels to see the list of available models

不是这个问题时 / 边界

The model name is in your ListModels output and you still get 404. (our inference: then check that you're calling the right API version, v1 or v1beta, and the right method.) (our inference: an alias like -latest can change behavior when Google moves it; pin a specific name in production.)

根因

The model name was retired or renamed. Old names stop resolving on the API.

修复步骤

1. List the models your key can use: GET https://generativelanguage.googleapis.com/v1beta/models?key=<YOUR_KEY>.
2. Change the model name to a current one from that list, for example gemini-2.5-flash, or the alias gemini-flash-latest (top answer). For embeddings, a commenter switched text-embedding-004 to gemini-embedding-001.
3. If you're on the deprecated SDKs (google.generativeai for Python, @google/generative-ai for Node), move to the supported ones (google-genai, @google/genai).

如何验证

The same call returns content instead of 404.