Where LiteLLM fits
LiteLLM is the strongest choice when you want an open-source unifying layer you fully control. If you already hold provider contracts, need to reach niche or self-hosted models, want to run the gateway inside your own VPC for compliance, and value a Python-native SDK plus a language-agnostic proxy with budgets and virtual keys, LiteLLM is built for exactly that. There is no inference lock-in — it simply standardises how your code calls whatever providers you configure.
Where Kunavo wins
Kunavo removes both the provider accounts and the ops. There is nothing to deploy: top up one wallet and call Claude, Gemini, GPT, plus image, video and audio, at 30–70% under each provider's list price depending on the model — no Anthropic Console, no Google Cloud project, no proxy to host and monitor, one invoice. Generation endpoints are first-class, the native Anthropic Messages API is on the same key, and checkout is Stripe-native with local methods. Where LiteLLM gives you a gateway to run, Kunavo gives you a gateway to call.
How to pick (or use both)
Pick LiteLLM if you want an open-source, self-hosted gateway over provider keys you already manage, or need niche/self-hosted model coverage inside your own infrastructure. Pick Kunavo if you want zero-ops hosted inference on one prepaid wallet at 30–70% under list, first-class multimodal, and one invoice. They compose: register Kunavo as an OpenAI-compatible provider in LiteLLM (base URL https://api.kunavo.com/v1) to keep LiteLLM's routing and budgets while your tokens run through Kunavo's discounted catalog.