docker本地部署接入Ollama视觉模型gemma3:12b分析图片报错

[Ivory-Chan]

Describe the bug
docker本地部署接入Ollama视觉模型gemma3:12b分析图片报错，报错信息如下：
Error: 400 invalid image input
at APIError.generate (/app/node_modules/.pnpm/[email protected][email protected][email protected]/node_modules/openai/error.js:45:20)
at OpenAI.makeStatusError (/app/node_modules/.pnpm/[email protected][email protected][email protected]/node_modules/openai/core.js:299:33)
at OpenAI.makeRequest (/app/node_modules/.pnpm/[email protected][email protected][email protected]/node_modules/openai/core.js:343:30)
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at async /app/node_modules/.pnpm/@langchain+openai@0.4.3_@langchain[email protected][email protected]/node_modules/@langchain/openai/dist/chat_models.cjs:1519:29
at async RetryOperation._fn (/app/node_modules/.pnpm/[email protected]/node_modules/p-retry/index.js:50:12)

To Reproduce
Steps to reproduce the behavior:

1. 上传图片节点（图片格式 png，大小800kb左右）
2. 使用AI问答节点并将图片节点作为上下文

Screenshots

Desktop (please complete the following information):

• OS: Mac OS M1 Pro
• Browser edge
• Version 15.3.2

Additional context
docker desktop： 4.32.2

[pftom]

@Ivory-Chan 看起来可能是 image 格式不兼容，你是 Openrouter 嘛

[Ivory-Chan]

本地运行的ollama 0.6.2