Skip to main content

MCP server for vision capabilities with automatic model fallback via ModelScope API

Project description

vision-mcp-server-pro

MCP server for vision capabilities with automatic model fallback via ModelScope API.

Features

  • Analyze images via URL or local file path
  • Automatic model fallback: when the primary model hits rate limits, it automatically tries the next model
  • Configurable model list via environment variables

Default Model List (fallback order)

  1. Qwen/Qwen3.5-397B-A17B
  2. Qwen/Qwen3-VL-235B-A22B-Instruct
  3. moonshotai/Kimi-K2.5
  4. Qwen/Qwen3.5-122B-A10B

Usage

Add to your MCP client (e.g., Claude Code) using uvx (no global install needed):

claude mcp add vision-mcp-server-pro --scope user \
  -e MODELSCOPE_TOKEN=your_token_here \
  -- uvx vision-mcp-server-pro

Environment Variables

使用前,需要设置以下环境变量:

Variable Required Description
MODELSCOPE_TOKEN Yes 魔搭社区的 API 密钥
MODELSCOPE_MODEL No 首选模型(默认使用回落列表中的第一个)
MODELSCOPE_FALLBACK_MODELS No 自定义回落模型列表,逗号分隔(覆盖默认列表)

获取 MODELSCOPE_TOKEN

访问 魔搭社区 → 个人中心 → API令牌

Specify Primary Model

claude mcp add vision-mcp-server-pro --scope user \
  -e MODELSCOPE_TOKEN=your_token_here \
  -e MODELSCOPE_MODEL=Qwen/Qwen3-VL-235B-A22B-Instruct \
  -- uvx vision-mcp-server-pro

Custom Fallback Models

claude mcp add vision-mcp-server-pro --scope user \
  -e MODELSCOPE_TOKEN=your_token_here \
  -e MODELSCOPE_FALLBACK_MODELS="Qwen/Qwen3.5-397B-A17B,Qwen/Qwen3-VL-235B-A22B-Instruct" \
  -- uvx vision-mcp-server-pro

Tool

The server provides an analyze_image tool:

  • image (required): Image URL or local file path
  • prompt (optional): Question or analysis request (default: "请描述这张图片的内容")

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

vision_mcp_server_pro-1.1.0.tar.gz (5.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

vision_mcp_server_pro-1.1.0-py3-none-any.whl (5.8 kB view details)

Uploaded Python 3

File details

Details for the file vision_mcp_server_pro-1.1.0.tar.gz.

File metadata

  • Download URL: vision_mcp_server_pro-1.1.0.tar.gz
  • Upload date:
  • Size: 5.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for vision_mcp_server_pro-1.1.0.tar.gz
Algorithm Hash digest
SHA256 207765a534bc8f40b3a94b67a0aba0a762393cc62ae1d933102623515cfc2ee5
MD5 f52b7c37507f67cacbcf42fca7865ec2
BLAKE2b-256 1e0a65c27f3c3e3291488cfeb5a9ca896e4feae742b056e63026b1bb0734d0c4

See more details on using hashes here.

Provenance

The following attestation bundles were made for vision_mcp_server_pro-1.1.0.tar.gz:

Publisher: publish.yml on lyf534631748/vision-mcp-server-pro

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file vision_mcp_server_pro-1.1.0-py3-none-any.whl.

File metadata

File hashes

Hashes for vision_mcp_server_pro-1.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 dc1b56de940942ce51b16b9d78aa07d2a98187f6c298f220433d264f12a1f5b0
MD5 1386477269e83eb482321e017a4652c4
BLAKE2b-256 e956180686d826ad995953135955dbcb4fe641c7099a79934a34e65ad1cff723

See more details on using hashes here.

Provenance

The following attestation bundles were made for vision_mcp_server_pro-1.1.0-py3-none-any.whl:

Publisher: publish.yml on lyf534631748/vision-mcp-server-pro

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page