Repository navigation
Gemini 3 Pro "Function call is missing a thought_signature in functionCall parts" #8785
Description
Before submitting your bug report
- I've tried using the "Ask AI" feature on the Continue docs site to see if the docs have an answer
- I'm not able to find a related conversation on GitHub discussions that reports the same bug
- I'm not able to find an open issue that reports the same bug
- I've seen the troubleshooting guide on the Continue Docs
Relevant environment info
- OS:macOS
- Continue version: 1.2.10
- IDE version: 1.106.0
- Model: gemini-3-pro-preview
- config:
- name: Gemini 3 Pro
provider: gemini
model: gemini-3-pro-preview
apiKey: ${{ secrets.GEMINI_API_KEY }}
roles:
- chat
- edit
- apply
defaultCompletionOptions:
contextLength: 1048576
maxTokens: 65536
capabilities:
- tool_use
- image_input
OR link to agent in Continue hub:Description
Tool use results in the following error:
[{ "error": { "code": 400, "message": "Function call is missing a thought_signature in functionCall parts. This is required for tools to work correctly, and missing thought_signature may lead to degraded model performance. Additional data, function call default_api:grep_search , position 3. please refer to https://ai.google.dev/gemini-api/docs/thought-signatures for more details.", "status": "INVALID_ARGUMENT" } } ]
To reproduce
No response
Log output
[@continuedev] error: HTTP 400 Bad Request from https://generativelanguage.googleapis.com/v1beta/models/gemini-3-pro-preview:streamGenerateContent
[{
"error": {
"code": 400,
"message": "Function call is missing a thought_signature in functionCall parts. This is required for tools to work correctly, and missing thought_signature may lead to degraded model performance. Additional data, function call `default_api:grep_search` , position 3. please refer to https://ai.google.dev/gemini-api/docs/thought-signatures for more details.",
"status": "INVALID_ARGUMENT"
}
}
] {"context":"llm_fetch","url":"https://generativelanguage.googleapis.com/v1beta/models/gemini-3-pro-preview:streamGenerateContent","method":"POST","model":"gemini-3-pro-preview","provider":"gemini"}
[@continuedev] error: HTTP 400 Bad Request from https://generativelanguage.googleapis.com/v1beta/models/gemini-3-pro-preview:streamGenerateContent
[{
"error": {
"code": 400,
"message": "Function call is missing a thought_signature in functionCall parts. This is required for tools to work correctly, and missing thought_signature may lead to degraded model performance. Additional data, function call `default_api:grep_search` , position 3. please refer to https://ai.google.dev/gemini-api/docs/thought-signatures for more details.",
"status": "INVALID_ARGUMENT"
}
}
] {"context":"llm_stream_chat","model":"gemini-3-pro-preview","provider":"gemini","useOpenAIAdapter":false,"streamEnabled":true,"templateMessages":false}Reacted by mboulton-fathom, nansenli, Star Brilliant, Iven S., Barret Miller, Andrii F and Vlad Baranchuk
Activity
murphyXu commentedon Nov 19, 2025
I am encountering a 400 INVALID_ARGUMENT error in agent mode when using Gemini 3.0 Pro via both Vertex AI and OpenRouter. According to the Google Vertex documentation, Gemini 3.0 Pro requires 'Thought Signatures'. I noticed that Cline has already implemented support for this. I look forward to seeing support for Thought Signatures in Continue for Gemini 3.0 Pro.
doc: https://docs.cloud.google.com/vertex-ai/generative-ai/docs/start/get-started-with-gemini-3#thought_signatures
cline ：cline/cline#7536
jmwang326 commentedon Nov 21, 2025
Same issue, although for the first few hours of gemini 3.0 existence, this wasn't an issue.
bpafoshizle commentedon Nov 23, 2025
Same issue for me. Is anyone working on this?
yashoswalyo commentedon Dec 22, 2025
I am also facing same issue, using latest version