Schema | llm-vision-mcp

llm-vision-mcp

Describes the environment variables required to run the server.

Name	Required	Description	Default
`TOOLS`	No	Comma-separated whitelist of tools to enable (e.g., 'analyze_image,extract_text'). Leave empty to enable all.
`API_KEY`	Yes	Provider API key. For ModelScope tokens, remove the 'ms-' prefix.
`BASE_URL`	No	OpenAI-compatible API endpoint.	https://api-inference.modelscope.cn/v1
`MAX_TOKENS`	No	Maximum number of tokens to generate (positive integer).	32768
`TEMPERATURE`	No	Sampling temperature (e.g., '0' or '0.5').	0.7
`VISION_MODEL`	No	Vision model name.	Qwen/Qwen3-VL-8B-Instruct

Features and capabilities supported by this server

Capability	Details
`tools`	{ "listChanged": true }

Functions exposed to the LLM to take actions

Name	Description
analyze_imageC	分析磁盘上的图片文件 / Analyze an image file on disk
extract_textB	从磁盘图片中提取文字(OCR) / OCR an image file on disk
describe_uiB	描述磁盘上 UI 图片文件 / Describe a UI screenshot file
diagnose_errorA	诊断磁盘上错误图片文件 / Diagnose an error screenshot file
understand_diagramA	解读流程图/架构图等图表 / Interpret a diagram image file
analyze_chartB	分析数据图表中的趋势和洞察 / Analyze a chart image file
code_from_screenshotB	从磁盘图片中提取代码 / Extract code from a screenshot file

Interactive templates invoked by user choice

Name	Description
No prompts

Contextual data attached and managed by the client

Name	Description
No resources

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/me9rez/llm-vision-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server