AI for everyone.

One API for leading models, with clear provider availability and request-level billing.

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.kineticrouter.com/v1",
    api_key=os.environ["KINETICROUTER_API_KEY"],
)

response = client.chat.completions.create(
    model="openai/gpt-6-astra",
    messages=[{"role": "user", "content": "Hello!"}],
)

print(response.choices[0].message.content)

38 modelsCurrent catalog
snapshot

57 docsStatus-marked
references

Connect Codex

1Set KINETICROUTER_API_KEY in your terminal environment
2Add the provider to ~/.codex/config.toml without replacing existing settings
3Start Codex with a model your key can access through Responses

# Add this provider alongside your existing configuration.
[model_providers.kineticrouter]
name = "kineticRouter"
base_url = "https://api.kineticrouter.com/v1"
env_key = "KINETICROUTER_API_KEY"
wire_api = "responses"

codex -c model_provider="kineticrouter" -m "YOUR_AVAILABLE_MODEL_ID"

Config~/.codex/config.tomlView setup guide

Popular reference prices

Compare kineticRouter prices with official reference rates. Discounts vary by provider.

View full model pricing

Anthropic: Claude Sonnet 5.5

anthropic/claude-sonnet-5-5 · Available

-85%

Per 1M tokenskineticRouterOfficial ref.

Input$0.3/M$2/M

Output$1.5/M$10/M

Cache read$0.015/M$0.1/M

OpenAI: GPT-6 Astra

openai/gpt-6-astra · Available

-80%

Per 1M tokenskineticRouterOfficial ref.

Input$2/M$10/M

Output$10/M$50/M

Cache read$0.2/M$1/M

Grok: Grok 4.7

grok/grok-4.7 · Available

-90%

Per 1M tokenskineticRouterOfficial ref.

Input$0.2/M$2/M

Output$0.6/M$6/M

Cache read——