ai-apiv2

Unify five OpenAI-compatible LLM APIs under a single interface.

Updated Apr 26, 2026
One-click install
npx skills add https://github.com/EndUser123/cc-marketplace --skill ai-apiv2
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-apiv2
Source: https://github.com/EndUser123/cc-marketplace/tree/main/plugins/cc-skills-ai-api/skills/ai-apiv2
Command: npx skills add https://github.com/EndUser123/cc-marketplace --skill ai-apiv2

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires openai, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Unified access for five OpenAI-compatible LLM providers under a single interface, simplifying integration and fallback logic.

Core Features & Use Cases

  • Unified OpenAI-compatible API across Chutes, OpenRouter, NVIDIA NIM, Gemini, and z.ai for multi-provider apps.
  • Streaming chat, code review support, model-fallback sequences, and cost optimization via provider selection.
  • Use cases include multi-model inference, resilience against provider outages, and rapid prototyping of multi-provider workflows.

Quick Start

Provide API keys for all five providers, initialize the OpenAI SDK with the provider-specific base_url, and POST to /v1/chat/completions while handling streaming responses.

Frequently Asked Questions about ai-apiv2

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I unify multiple OpenAI-compatible LLM APIs under a single interface?

To unify multiple OpenAI-compatible LLM APIs, this interface connects Chutes, OpenRouter, NVIDIA NIM, Gemini, and z.ai using provider-specific base URLs and the OpenAI SDK. You provide API keys via environment variables and route requests to the desired provider.

Can I implement provider fallbacks for LLM streaming chat using the OpenAI SDK?

Yes, you can implement provider fallbacks for LLM streaming chat. By configuring model-fallback sequences across the five providers, the system maintains resilience against provider outages while handling streaming responses via the OpenAI SDK.

How does multi-provider LLM access handle code review workflows?

Multi-provider LLM access supports code review workflows by routing review requests through a unified OpenAI-compatible API. You can select specific providers or use fallback sequences to ensure continuous code analysis during outages.

Do I need separate API keys for each LLM provider to use this unified interface?

Yes, you need separate API keys for each LLM provider. The interface uses environment-variable configuration to manage credentials for Chutes, OpenRouter, NVIDIA NIM, Gemini, and z.ai, ensuring secure access to all five providers.

What is the best way to optimize LLM costs across multiple providers?

The best way to optimize LLM costs is through provider selection and model-fallback sequences. This unified interface allows you to route requests to specific providers like Chutes or OpenRouter based on cost, switching automatically if a provider fails.

Does the unified LLM API work with the standard OpenAI SDK for streaming completions?

Yes, the unified LLM API works with the standard OpenAI SDK. You initialize the SDK with provider-specific base URLs and POST to the /v1/chat/completions endpoint, handling streaming responses exactly as you would with native OpenAI.