Gemini-2.0-flash rejects function schemas containing propertyNames, causing 400 Bad Request #8584
Description
Checked other resources
- This is a bug, not a usage question. For questions, please use the LangChain Forum (https://forum.langchain.com/).
- I added a very descriptive title to this issue.
- I searched the LangChain.js documentation with the integrated search.
- I used the GitHub search to find a similar question and didn't find it.
- I am sure that this is a bug in LangChain.js rather than my code.
- The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).
Example Code
import { ChatPromptTemplate } from '@langchain/core/prompts';
import { tool } from '@langchain/core/tools';
import { ChatGoogleGenerativeAI } from '@langchain/google-genai';
import { AgentExecutor, createToolCallingAgent } from 'langchain/agents';
import z from 'zod';
const MagicNumberSchema = z.object({
input: z.number().describe('Input number to apply the magic function to.'),
someVeryComplexField: z.record(z.string(), z.string()).optional().describe(
'A complex field that can hold various string key-value pairs.',
),
});
const prompt = ChatPromptTemplate.fromMessages([
['system', 'You are a helpful assistant'],
['placeholder', '{chat_history}'],
['human', '{input}'],
['placeholder', '{agent_scratchpad}'],
]);
const llm = new ChatGoogleGenerativeAI({
model: 'gemini-2.0-flash',
apiKey: Deno.env.get('GOOGLE_GENERATIVE_AI_API_KEY'),
baseUrl: Deno.env.get('GOOGLE_GENERATIVE_AI_API_URL') || 'https://generativelanguage.googleapis.com',
streaming: false,
});
const magicTool = tool(
({ input }) => {
return `${input + 2}`;
},
{
name: 'magic_function',
description: 'Applies a magic function to an input.',
schema: MagicNumberSchema,
},
);
const agent = createToolCallingAgent({
llm,
tools: [magicTool],
prompt,
});
const agentExecutor = new AgentExecutor({
agent,
tools: [magicTool],
});
const response = await agentExecutor.invoke({
input: 'Hello, what is 2 + 2?',
});
console.log(response.output); // Should log the result of the magic function applied to the input instead it throws an errorError Message and Stack Trace (if applicable)
error: Uncaught (in promise) Error: [GoogleGenerativeAI Error]: Error fetching from https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash:streamGenerateContent?alt=sse: [400 Bad Request] Invalid JSON payload received. Unknown name "propertyNames" at 'tools[0].function_declarations[0].parameters.properties[1].value': Cannot find field. [{"@type":"type.googleapis.com/google.rpc.BadRequest","fieldViolations":[{"field":"tools[0].function_declarations[0].parameters.properties[1].value","description":"Invalid JSON payload received. Unknown name \"propertyNames\" at 'tools[0].function_declarations[0].parameters.properties[1].value': Cannot find field."}]}]
at handleResponseNotOk (/node_modules/.deno/@google+generative-ai@0.24.1/node_modules/@google/generative-ai/dist/index.mjs:432:11)
at eventLoopTick (ext:core/01_core.js:178:7)
at async makeRequest (/node_modules/.deno/@google+generative-ai@0.24.1/node_modules/@google/generative-ai/dist/index.mjs:401:9)
at async generateContentStream (/node_modules/.deno/@google+generative-ai@0.24.1/node_modules/@google/generative-ai/dist/index.mjs:860:22)
at async /node_modules/.deno/@langchain+google-genai@0.2.16/node_modules/@langchain/google-genai/dist/chat_models.js:690:32
at async RetryOperation._fn (/node_modules/.deno/p-retry@4.6.2/node_modules/p-retry/index.js:50:12)
Description
When using @langchain/google-genai with Gemini-2.0-flash, any Zod schema that includes dynamic keys (e.g., z.record(...)) generates a JSON schema containing the propertyNames keyword. The Gemini API strictly validates the schema and returns a 400 Bad Request with the message.
System Info
pnpm info langchain
langchain@0.3.30 | MIT | deps: 11 | versions: 325
Typescript bindings for langchain
https://github.com/langchain-ai/langchainjs/tree/main/langchain/
keywords: llm, ai, gpt3, chain, prompt, prompt engineering, chatgpt, machine learning, ml, openai, embeddings, vectorstores
dist
.tarball: https://registry.npmjs.org/langchain/-/langchain-0.3.30.tgz
.shasum: 9fc0467007009dc5e799d1f8fb9dd1137654e387
.integrity: sha512-UyVsfwHDpHbrnWrjWuhJHqi8Non+Zcsf2kdpDTqyJF8NXrHBOpjdHT5LvPuW9fnE7miDTWf5mLcrWAGZgcrznQ==
.unpackedSize: 2.9 MB
dependencies:
@langchain/openai: >=0.1.0 <0.7.0 js-yaml: ^4.1.0 openapi-types: ^12.1.3 yaml: ^2.2.1
@langchain/textsplitters: >=0.0.0 <0.2.0 jsonpointer: ^5.0.1 p-retry: 4 zod: ^3.25.32
js-tiktoken: ^1.0.12 langsmith: ^0.3.33 uuid: ^10.0.0
maintainers:
- nfcampos <nuno@boringbits.io>
- jacoblee93 <jacoblee93@gmail.com>
- andrewnguonly <andrewnguonly@gmail.com>
- davidduong <david@duong.cz>
- hntrl <hunter@hntrl.io>
- hwchase17 <hw.chase.17@gmail.com>
- basproul <braceasproul@gmail.com>
dist-tags:
latest: 0.3.30 tag-for-publishing-older-releases: 0.2.20 next: 0.3.2-rc.0
published a week ago by davidduong <david@duong.cz>