relayswitch

Python client for RelaySwitch: one API key for Claude, GPT, Gemini, DeepSeek, Qwen, Kimi, GLM and more, built for coding agents and paid by card or crypto. The API speaks the OpenAI and Anthropic formats, so you keep the official SDKs.

pip install "relayswitch[openai]"      # or [anthropic], or [all]
export RELAYSWITCH_API_KEY="..."

Get a key at app.relayswitch.io/console. Credit is prepaid from $10, with no subscription.

OpenAI SDK

import relayswitch

client = relayswitch.openai_client()
reply = client.chat.completions.create(
    model="claude-opus-5-5",
    messages=[{"role": "user", "content": "Say hi in five words."}],
)
print(reply.choices[0].message.content)

Anthropic SDK

import relayswitch

client = relayswitch.anthropic_client()
msg = client.messages.create(
    model="claude-sonnet-5-5",
    max_tokens=200,
    messages=[{"role": "user", "content": "Say hi in five words."}],
)
print(msg.content[0].text)

Prices, no key needed

import relayswitch

for m in relayswitch.models():
    p = m["pricing"]
    print(f'{m["id"]:<40} ${p["input"]}/${p["output"]} per 1M tokens')

Without this package

Set OPENAI_BASE_URL or ANTHROPIC_BASE_URL to https://api.relayswitch.io/v1 and use your key.

• Models and prices: relayswitch.io/models
• Daily real-model check: relayswitch.io/proof
• Setup guides: relayswitch.io/guides

Run by Uplink Switch LLC.

Sister sites

The same gateway, sold each on its own key and console:

• UncensoredLLM: the uncensored and abliterated builds as an API, priced per token.
• Unrestrictable: uncensored AI chat, images and video in the browser, no key needed.
• PentestLLM: uncensored AI for hacking and pentesting.
• UplinkSwitch: the gateway itself: one key and one base URL for every model.

All of them: UplinkSwitch for every model, UncensoredLLM for abliterated builds, Unrestrictable for chat in the browser, PentestLLM for security work, RelaySwitch for coding agents.