Guides / Switch to wylon

Switch to wylon

wylon is compatible with both the OpenAI and Anthropic client SDKs. Whether you’re migrating from GPT or Claude, the change takes three small configuration updates — no new libraries, no request schema to relearn.

Choose your SDK

Select the SDK you’re currently using to see the relevant migration steps.

What changes — OpenAI SDK

Only three values in your existing OpenAI code need to be updated. Everything else — request parameters, response schema, streaming, tool use — stays identical.

Setting	OpenAI	wylon
`base_url`	`https://api.openai.com/v1`	`https://api.wylon.cn/v1`
`api_key`	`OPENAI_API_KEY`	`WYLON_API_KEY`
`model`	`gpt-4o`, `gpt-4o-mini`, …	`moonshotai/kimi-k2.5`, and more

Side-by-side comparison

The diff below shows a minimal OpenAI script and its wylon equivalent. Only the highlighted lines change.

from openai import OpenAI
import os

client = OpenAI(
    api_key=os.environ["OPENAI_API_KEY"],
)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user",   "content": "Explain KV cache in one paragraph."},
    ],
    temperature=0.6,
    max_tokens=512,
)

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

from openai import OpenAI
import os

client = OpenAI(
    api_key=os.environ["WYLON_API_KEY"],         # ← changed
    base_url="https://api.wylon.cn/v1",         # ← added
)

response = client.chat.completions.create(
    model="moonshotai/kimi-k2.5",               # ← changed
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user",   "content": "Explain KV cache in one paragraph."},
    ],
    temperature=0.6,
    max_tokens=512,
)

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

Full example — OpenAI SDK

A complete, runnable script. Set WYLON_API_KEY and run it directly.

from openai import OpenAI
import os

client = OpenAI(
    api_key=os.environ["WYLON_API_KEY"],
    base_url="https://api.wylon.cn/v1",
)

response = client.chat.completions.create(
    model="moonshotai/kimi-k2.5",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user",   "content": "What are the benefits of open-source LLMs?"},
    ],
    temperature=0.7,
    max_tokens=512,
)

print(response.choices[0].message.content)
print(f"\nTokens used: {response.usage.total_tokens}")

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.WYLON_API_KEY,
  baseURL: "https://api.wylon.cn/v1",
});

const response = await client.chat.completions.create({
  model: "moonshotai/kimi-k2.5",
  messages: [
    { role: "system", content: "You are a helpful assistant." },
    { role: "user",   content: "What are the benefits of open-source LLMs?" },
  ],
  temperature: 0.7,
  max_tokens: 512,
});

console.log(response.choices[0].message.content);
console.log(`\nTokens used: ${response.usage.total_tokens}`);

curl https://api.wylon.cn/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $WYLON_API_KEY" \
  -d '{
    "model": "moonshotai/kimi-k2.5",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user",   "content": "What are the benefits of open-source LLMs?"}
    ],
    "temperature": 0.7,
    "max_tokens": 512
  }'

What changes — Anthropic SDK

Three values need updating. The Anthropic SDK’s messages.create interface, streaming, and tool use all continue to work without further changes.

Setting	Anthropic	wylon
`base_url`	(default)	`https://api.wylon.cn`
`api_key`	`ANTHROPIC_API_KEY`	`WYLON_API_KEY`
`model`	`claude-opus-4-5`, `claude-sonnet-4-6`, …	`moonshotai/kimi-k2.5`, and more

info

Note on base_url. The Anthropic SDK appends /v1/messages automatically. Pass https://api.wylon.cn without a trailing path so the SDK constructs the correct endpoint.

Side-by-side comparison

A minimal Anthropic script and its wylon equivalent. Only the highlighted lines change.

import anthropic
import os

client = anthropic.Anthropic(
    api_key=os.environ["ANTHROPIC_API_KEY"],
)

message = client.messages.create(
    model="claude-opus-4-5",
    max_tokens=512,
    system="You are a helpful assistant.",
    messages=[
        {"role": "user", "content": "Explain KV cache in one paragraph."},
    ],
)

print(message.content[0].text)

import anthropic
import os

client = anthropic.Anthropic(
    api_key=os.environ["WYLON_API_KEY"],          # ← changed
    base_url="https://api.wylon.cn",             # ← added
)

message = client.messages.create(
    model="moonshotai/kimi-k2.5",               # ← changed
    max_tokens=512,
    system="You are a helpful assistant.",
    messages=[
        {"role": "user", "content": "Explain KV cache in one paragraph."},
    ],
)

print(message.content[0].text)

Full example — Anthropic SDK

A complete, runnable script. Set WYLON_API_KEY and run it directly.

import anthropic
import os

client = anthropic.Anthropic(
    api_key=os.environ["WYLON_API_KEY"],
    base_url="https://api.wylon.cn",
)

message = client.messages.create(
    model="moonshotai/kimi-k2.5",
    max_tokens=512,
    system="You are a helpful assistant.",
    messages=[
        {"role": "user", "content": "What are the benefits of open-source LLMs?"},
    ],
)

print(message.content[0].text)
print(f"\nInput tokens:  {message.usage.input_tokens}")
print(f"Output tokens: {message.usage.output_tokens}")

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  apiKey: process.env.WYLON_API_KEY,
  baseURL: "https://api.wylon.cn",
});

const message = await client.messages.create({
  model: "moonshotai/kimi-k2.5",
  max_tokens: 512,
  system: "You are a helpful assistant.",
  messages: [
    { role: "user", content: "What are the benefits of open-source LLMs?" },
  ],
});

console.log(message.content[0].text);
console.log(`\nInput tokens:  ${message.usage.input_tokens}`);
console.log(`Output tokens: ${message.usage.output_tokens}`);

curl https://api.wylon.cn/v1/messages \
  -H "Content-Type: application/json" \
  -H "x-api-key: $WYLON_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "moonshotai/kimi-k2.5",
    "max_tokens": 512,
    "system": "You are a helpful assistant.",
    "messages": [
      {"role": "user", "content": "What are the benefits of open-source LLMs?"}
    ]
  }'

Set your environment variable

Generate a wylon API key from Account Settings → API keys in the dashboard, then export it to your shell. The same key works with both the OpenAI and Anthropic SDKs.

# Add to ~/.bashrc or ~/.profile
export WYLON_API_KEY="wl-••••••••••••••••••••••••••••••••"

# Add to ~/.zshrc
export WYLON_API_KEY="wl-••••••••••••••••••••••••••••••••"

# PowerShell — persistent for the current user
[Environment]::SetEnvironmentVariable("WYLON_API_KEY", "wl-••••••••••••••••••••••••••••••••", "User")

key

Keep your key secret. Never commit API keys to source control or ship them in client-side bundles. Use a secret manager or a server-side proxy for production workloads.

What stays the same

wylon mirrors both API surfaces end-to-end. The following capabilities require zero changes to your existing code, regardless of which SDK you use.

Feature	OpenAI SDK	Anthropic SDK
Request parameters (`temperature`, `max_tokens`, `top_p`, …)	Supported	Supported
Streaming via server-sent events	Supported	Supported
Function calling & tool use	Supported	Supported
Structured output / JSON mode	Supported	Supported
Multi-turn conversation history	Supported	Supported
LangChain, LlamaIndex, LiteLLM integrations	Supported	Supported

info

Model IDs are different. Neither OpenAI (gpt-4o) nor Anthropic (claude-opus-4-5) model names exist on wylon. Use a wylon model ID such as moonshotai/kimi-k2.5. Browse all available models on the Models page.

Next steps

You’re migrated. Here’s what to explore next.

Browse models arrow_forward

Compare frontier open-source models by capability, context length, and per-token price.

Function calling & tools arrow_forward

Teach models to invoke your code, query databases, and orchestrate tool chains reliably.

Streaming responses arrow_forward

Deliver token-by-token output with server-sent events using the same stream API you know.

Structured output & JSON arrow_forward

Constrain generations to JSON Schema. Parse confidently — no regex, no retries.