i was using vscode insiders, and trying BYOM, like to use gemini models via openai compaitable api. i enabled function calls and all. it generates responses after thinking without edits.
these are my configs.

the error says "Function call is missing a thought_signature in functionCall parts" which is needed for gemini
i think we have the docs here.

i was using vscode insiders, and trying BYOM, like to use gemini models via openai compaitable api. i enabled function calls and all. it generates responses after thinking without edits.
these are my configs.

the error says "Function call is missing a thought_signature in functionCall parts" which is needed for gemini
i think we have the docs here.