frontier

OpenAI

Reference frontier stack. Strong tool-use and structured outputs.

live

Uptime (30d)

99.94%

Median tps

Median TTFT

380 ms

Price index

1.6×

vs. market median

Overview

HQ: San Francisco, US
Founded: 2015
Models: 14
Modalities: text, vision, audio
Regions: us-east, us-west, eu-west

Strengths

reasoningtool-usestructured-output

Route via llmcloud

curl https://api.llmcloud.ai/v1/chat/completions \
  -H "Authorization: Bearer $LLMCLOUD_KEY" \
  -H "X-Provider: openai" \
  -d '{
    "model": "auto:quality",
    "messages": [{"role":"user","content":"hi"}]
  }'

Models served

gpt-5.5

q 93.874 tps$12.5/1M

gpt-5

q 92.478 tps$10/1M

gpt-5-codex

q 91.296 tps$4.5/1M

q 90.162 tps$10/1M

gpt-4.1

q 88.292 tps$8/1M

o4-mini

q 87.588 tps$1.1/1M

gpt-4o

q 86.1105 tps$5/1M

gpt-5.5-mini

q 85.3210 tps$0.4/1M

o3-mini

q 84.492 tps$1.1/1M

gpt-5-mini

q 84.1215 tps$0.35/1M

gpt-4.1-mini

q 82.6200 tps$0.4/1M

gpt-5.5-nano

q 78.4320 tps$0.1/1M

gpt-4o-mini

q 78.2240 tps$0.15/1M

gpt-4.1-nano

q 76.9300 tps$0.1/1M