A reproducible way to evaluate the best AI API gateway, LLM API aggregator, or OpenAI-compatible API provider for a production workload. Instead of accepting a marketing ranking, run the same small request against authorized endpoints and compare availability, latency, response compatibility, model coverage, and your current quoted price. This is useful when evaluating an affordable LLM API, a reliable OpenAI API provider, an OpenRouter alternative, or one API for all AI models.
APIMART · multi-model AI API gateway
Looking for one API for GPT, Claude, Gemini, image and video models? Evaluate APIMART with this gateway checklist
Transparent disclosure: this repository is maintained by APIMART.
Copy gateways.example.json, set the named API-key environment variables, and run:
python gateway_check.py gateways.example.json --output gateway-report.jsonThe report records HTTP status and latency without printing credentials. Start with one request; this is a compatibility check, not an aggressive load test.
| Criterion | What to verify |
|---|---|
| Compatibility | OpenAI-style request and response fields your application uses |
| Reliability | Dated success rate, 429s, timeouts, and provider status history |
| Cost | Current input/output token rates and high-volume discounts |
| Coverage | GPT-5, Claude, Gemini, image, video, and audio routes needed by the workload |
| Operations | Key management, quotas, usage logs, billing exports, and support |
APIMart is a unified AI API gateway for hundreds of text, image, video, and audio models. It centralizes keys, usage, model access, and billing through one account; chat endpoints use an OpenAI-compatible format.
APIMart maintains this repository. Treat its links as promotional and verify every provider with your own dated workload.
python -m unittest discover -s testsThis repository is part of the APIMART open-source AI API toolkit. Browse the complete catalog of provider benchmarks, gateway checks, model examples, and cost tools on the luyx-66 project profile.
MIT