Skip to main content
Glama

glm-vision-mcp

An MCP server that wraps the Zhipu GLM-4.6V-Flash (free vision model), exposing an analyze_image tool to any MCP client, with support for single/multi-image analysis, OCR, and multi-image comparison.

Features

Capability

Description

Image analysis

Local paths / http(s) URLs / base64 data URIs are all supported, automatically converted to data URIs

Multi-image comparison

Pass multiple images in a single call and compare them according to the prompt

Rate-limit resilience

429 / 1302 / 1305 exponential backoff retry → multi-key rotation → fallback to backup model glm-4.1v-thinking-flash

Configuration self-check

The check_config tool verifies Key/model/endpoint without leaking the Key

Related MCP server: vision-mcp

Requirements

  • Python >= 3.10

  • Zhipu Open Platform API Key (https://open.bigmodel.cn/usercenter/apikeys), glm-4.6v-flash is free

  • To further reduce the chance of rate limiting, you can register multiple accounts and obtain multiple Keys, separated by commas in the configuration

Installation

cd glm-vision-mcp
python -m venv .venv
.venv\Scripts\pip install -r requirements.txt

Startup

# stdio 模式(MCP 客户端默认方式)
$env:ZHIPU_API_KEY = "你的Key"
.venv\Scripts\python server.py

# SSE 调试模式(无鉴权,仅限本机)
.venv\Scripts\python server.py --sse 8090

Client Configuration

Codex (~/.codex/config.toml)

[mcp_servers.glm-vision]
command = "C:\\绝对路径\\glm-vision-mcp\\.venv\\Scripts\\python.exe"
args = ["C:\\绝对路径\\glm-vision-mcp\\server.py"]

[mcp_servers.glm-vision.env]
ZHIPU_API_KEY = "你的Key"
# GLM_VISION_MODELS = "glm-4.6v-flash"
# GLM_API_BASE = "https://open.bigmodel.cn/api/paas/v4/chat/completions"

Claude Desktop (claude_desktop_config.json)

{
  "mcpServers": {
    "glm-vision": {
      "command": "C:\\绝对路径\\glm-vision-mcp\\.venv\\Scripts\\python.exe",
      "args": ["C:\\绝对路径\\glm-vision-mcp\\server.py"],
      "env": { "ZHIPU_API_KEY": "你的Key" }
    }
  }
}

Tool Interface

analyze_image(images, prompt, temperature, max_tokens, thinking)

Parameter

Type

Required

Description

images

string[]

Yes

Local path / http(s) URL / data URI

prompt

string

No

Analysis request, default "Please describe the content of this image in detail"

temperature

number

No

0.0~1.0, default 0.7

max_tokens

integer

No

Maximum output tokens, default 2048

thinking

boolean

No

Deep thinking mode, default false

Environment Variables

Variable

Required

Description

ZHIPU_API_KEY

Yes

Zhipu API Key, comma-separated for multi-key rotation

GLM_VISION_MODELS

No

Model priority, comma-separated, default glm-4.6v-flash,glm-4.1v-thinking-flash

GLM_API_BASE

No

Override API endpoint

Notes

  • Local images must be ≤ 10MB per image, supported formats: jpg/jpeg/png/webp/gif/bmp

  • Free models may be rate-limited during peak hours; an error is only raised when all Keys + all models are rate-limited. Wait 15~30 seconds and retry at off-peak times

  • Non-rate-limit errors such as 401 are not retried or downgraded; they are returned directly for easier troubleshooting

Verification

# 离线检查(不联网)
.venv\Scripts\python test_smoke.py

# 联网冒烟:MCP 握手 + analyze_image 真实调用
$env:ZHIPU_API_KEY = "你的Key"
.venv\Scripts\python test_smoke.py --live

Available Tools

2 tools
analyze_imageA

调用智谱 GLM-4.6V-Flash 分析一张或多张图片。

适用:图片描述、OCR 文字提取、表格/图表解析、UI 截图理解、多图对比等。

Args: images: 图片列表,每项是本地文件路径、http(s):// URL 或 base64 data URI。 prompt: 对图片的提问或分析指令,如 “提取图中所有文字,保留排版顺序”。 temperature: 采样温度 0.0~1.0,越小越确定,默认 0.7。 max_tokens: 最大输出 token 数,默认 2048。 thinking: 是否开启深度思考(返回附带思考过程),默认 False。

Returns: dict,含 model / content(/ reasoning / elapsed_ms)字段;失败抛异常。

Inputs: images (required), prompt, thinking, max_tokens, +1 more.

check_configA

检查 MCP 配置是否就绪(不泄露 Key 本体)。

Inputs: no named parameters.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.2/5.0

    Scored across 2 tools

    Disambiguation5/5

    The two tools are completely distinct: analyze_image handles all image analysis tasks while check_config verifies setup. There is zero overlap or ambiguity in their purposes.

    Naming Consistency5/5

    Both tools follow the consistent verb_noun snake_case convention—analyze_image and check_config. The pattern is clean and predictable, matching the higher-scoring examples.

    Tool Count3/5

    With only 2 tools in a server, the count falls into the 'feels thin' category. While the core vision analysis capability is covered by a single tool, the overall server feels minimal and could reasonably be expected to have a few more supporting tools to feel substantial.

    Completeness4/5

    The single analysis tool fully covers the core domain of image description, OCR, and chart parsing—no obvious dead ends for standard usage. The config check provides useful operational support. Minor gaps like batch processing or model inquiry are absent but not critical for the stated scope.

    Maintenance

    ActivityInactive
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      C
      maintenance
      Multi-model vision understanding MCP server that provides unified image analysis for AI assistants without native vision, supporting models like GLM-4.6V, DeepSeek-OCR, Qwen3-VL-Flash, and more.
      1
      512 npm
      116
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      MCP server for analyzing images using multiple vision LLM providers (OpenCode, OpenAI, Anthropic, Google, and custom OpenAI-compatible endpoints). Provides tools to analyze single or multiple images, list providers, and test vision capabilities.
      MIT