shared-vlm-routing

Community

Fail-safe VLM routing for accurate extractions.

AuthorAKCqhzdy
Version1.0.0
Installs0

System Documentation

What problem does it solve?

VLM provider routing with automatic failover across Gemini 3.1 Pro, Kimi K2.5, and ALIYUN to ensure reliable, cost-aware access to OpenAI-compatible models within grading pipelines.

Core Features & Use Cases

  • Multi-provider routing with sequential fallbacks and logging of provider switches to surface usage distribution.
  • OpenAI-compatible client usage via Vertex AI endpoint; supports parallel or sequential calls as needed to avoid rate limits.
  • Integrated with grading workflows (grading-paper-extraction, grading-answer-extraction) and any future VLM-using skills.

Quick Start

Configure your VLM task to use Gemini as the primary provider and allow automatic fallback to Kimi and ALIYUN on failure.

Dependency Matrix

Required Modules

openaigoogle-auth

Components

scripts

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: shared-vlm-routing
Download link: https://github.com/AKCqhzdy/dse-subject-grading/archive/main.zip#shared-vlm-routing

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.