Chat Completions API
AI / MLOpenAI
GPT-4o and GPT-4-turbo language model inference for chat, completion, and function calling
Reported Latency
Typical performance from public reports -- not live measurements
Average: 850ms
p50
600ms
p95
2,200ms
p99
4,500ms
Reliability
99.81%observed uptime
Pricing
Free tier
None (credit-based)
Paid starts at
$0.50 / 1M input tokens (GPT-4o-mini)
Technical Details
Authentication
API Key
Rate limit
500 RPM (Tier 1)
Protocols
RESTWebSocket (Realtime)
SDKs
PythonNode.jsGo.NETJava
Regions
USEU
Last updated: 2026-04-12