OrcaRouter as an optional generation provider for awesome-gpt-image-2

The "Prompt as Code" framing is what makes awesome-gpt-image-2 valuable. The repo reverse-engineered 500+ community GPT-Image2 examples into structured, reusable templates, and shares one style library across the website and an agent skill for Claude Code, Codex, and Cursor. That moves image generation from one-off luck toward repeatable output that scripts and agents can depend on, which is the core problem for batch or template-driven workflows.

It is also not just a static gallery: the companion site runs live generation against a provider endpoint configured by an API key, so users are already operating a real generation workflow. For that community, an optional additional provider behind that step is genuinely useful.

Proposal: OrcaRouter as an optional provider

I'm an engineer on the OrcaRouter team. We maintain OrcaRouter, a gateway that exposes many chat, reasoning, image, and video models through a single OpenAI-compatible endpoint with standard API-key authentication. I'd like to propose adding OrcaRouter as an optional provider for this repo's generation workflow. It would be purely additive: no existing provider such as APIMart would be replaced or changed, and prompts, templates, credits, and the gallery stay untouched.

Why it fits this repo

Three capabilities matter most here:

• Provider failover and automatic routing, so live generation can fall back when the current upstream is busy or down.
• Multiple models behind one endpoint: the provider config already names a model and a base URL, so an OpenAI-compatible router widens the set of models reachable through the same calling style.
• Usage tracking: the repo already stores provider cost per task (provider_cost_usd); router-level reporting complements that for self-hosters.

Integration shape

OrcaRouter exposes an OpenAI-compatible API and uses standard API-key authentication, matching this repo's existing provider abstraction: a config module like shared/apimart.js plus an environment-variable API key that calls a /v1/... endpoint. The natural fit is a new optional provider module beside APIMart, selected per deployment through its own env key. This is a proposal only - no code has been written or tested, and I would wait for maintainers to confirm the approach.

Ecosystem and disclosure

OrcaRouter integrations are appearing in open-source projects such as Dify, promptfoo, and CAMEL; we track them at https://www.orcarouter.ai/built-with. Transparently, OrcaRouter also runs an optional open-source partner program where approved OSS projects can receive a 5% revenue share from usage attributed to their integration. That is not a condition for integrating, and I'm glad to follow this repo's disclosure and governance rules, including omitting the program.

Next step

Would you be open to this direction? If so, I'd be glad to submit an implementation PR for review once the integration point is agreed. Thanks again for making image prompts genuinely reusable.