ReliAPI

Reliability layer for AI APIs. Automatic retries, caching, and fallbacks. Never lose a request to rate limits or downtime again.

Request

POST /v1/chat/completions
429 Rate Limited

ReliAPI Response

CACHE HIT
200 OK (from cache)
Latency: 12ms

Smart Retries

Exponential backoff with jitter

Response Cache

Semantic caching for identical prompts

Fallback Chains

Automatic failover to backup providers

Specifications

Supported Providers

OpenAI, Anthropic, Cohere, custom endpoints

Pricing Details

$0.005 per proxied request. Cache hits are free. You pay for the underlying API separately.

Pricing

Pay per proxied request

0.1credits / request

≈ $0.0001 per request

Free tier: 1,000/day, 10,000/month • Cache hits free

Get Started

Try it now

curl -X POST https://reliapi.kikuai.dev/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-4", "messages": [{"role": "user", "content": "Hello"}]}'