Migrate from OpenRouter to TeamoRouter

Moving from OpenRouter takes three changes: the base URL, the API key, and the model ID (drop the vendor/ prefix). The rest of your OpenAI SDK, Anthropic SDK, Claude Code or Codex setup stays the same.

1. What to change

2. Model ID mapping

Drop the vendor/ prefix. Claude IDs also swap dots for hyphens: anthropic/claude-opus-5.5 becomes claude-opus-5-5. A vendor-prefixed ID returns 400 model_not_available.

The table lists the main models both services offer, generated from both public model lists on 2026-11-04. For everything else and for live prices, see the model directory or GET /v1/models.

• openrouter/auto and ~vendor/...-latest aliases have no equivalent. Pick a model ID from the table.
• Variant suffixes such as :nitro, :floor, :online and :batch aren't supported. Remove the suffix. Free models are separate IDs ending in -free; they're marked in the model directory.
• Claude Code's [1m] suffix (claude-opus-5-5[1m]) isn't needed. Use claude-opus-5-5.

3. Headers you can delete

OpenRouter reads HTTP-Referer, X-OpenRouter-Title (formerly X-Title), X-OpenRouter-Categories and X-OpenRouter-App-Visibility only to list your app in its public rankings. TeamoRouter doesn't use them. Delete them when you switch.

4. Code changes

Python (OpenAI SDK)

Before:

from openai import OpenAI

client = OpenAI(
    base_url="https://openrouter.ai/api/v1",
    api_key="sk-or-v1-xxxxxx",
)

resp = client.chat.completions.create(
    model="deepseek/deepseek-v4-pro",
    messages=[{"role": "user", "content": "Hello"}],
    extra_headers={"HTTP-Referer": "https://example.com", "X-OpenRouter-Title": "My App"},
    extra_body={"provider": {"order": ["deepseek"]}},
)
print(resp.choices[0].message.content)

After:

from openai import OpenAI

client = OpenAI(
    base_url="https://api.teamorouter.com/v1",
    api_key="sk-teamo-xxxxxx",
)

resp = client.chat.completions.create(
    model="deepseek-v4-pro",
    messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)

JavaScript / TypeScript (OpenAI SDK)

Before:

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://openrouter.ai/api/v1",
  apiKey: process.env.OPENROUTER_API_KEY,
  defaultHeaders: { "HTTP-Referer": "https://example.com", "X-OpenRouter-Title": "My App" },
});

const resp = await client.chat.completions.create({
  model: "deepseek/deepseek-v4-pro",
  messages: [{ role: "user", content: "Hello" }],
});

After:

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.teamorouter.com/v1",
  apiKey: process.env.TEAMOROUTER_API_KEY,
});

const resp = await client.chat.completions.create({
  model: "deepseek-v4-pro",
  messages: [{ role: "user", content: "Hello" }],
});

Claude models: use the Anthropic format

On OpenRouter you may call Claude through the OpenAI SDK as anthropic/claude-.... Use the Anthropic native protocol (/v1/messages) for Claude models whenever possible. Calling Claude through the OpenAI-compatible format can lose prompt cache, thinking, and other capabilities, increasing cost and reducing quality. It is only suitable for simple chat scenarios.

from anthropic import Anthropic

client = Anthropic(
    api_key="sk-teamo-xxxxxx",
    base_url="https://api.teamorouter.com",
)

resp = client.messages.create(
    model="claude-sonnet-5-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello"}],
)
print(resp.content[0].text)

Caching has to be enabled by the caller with cache_control. See the FAQ in the API integration guide.

Claude Code

Before:

export ANTHROPIC_BASE_URL="https://openrouter.ai/api"
export ANTHROPIC_AUTH_TOKEN="sk-or-v1-xxxxxx"
export ANTHROPIC_API_KEY=""
export ANTHROPIC_DEFAULT_OPUS_MODEL="~anthropic/claude-opus-latest[1m]"
export CLAUDE_CODE_ENABLE_GATEWAY_MODEL_DISCOVERY=1

After:

export ANTHROPIC_BASE_URL="https://api.teamorouter.com"
export ANTHROPIC_AUTH_TOKEN="sk-teamo-xxxxxx"
export ANTHROPIC_API_KEY=""
export ANTHROPIC_DEFAULT_OPUS_MODEL="claude-opus-5-5"

• Keep ANTHROPIC_API_KEY empty, as on OpenRouter. If it's set, Claude Code asks whether to use that key, defaults to No, and pressing Enter takes you to the official Anthropic login.
• Change every model variable (ANTHROPIC_MODEL, ANTHROPIC_DEFAULT_*_MODEL, CLAUDE_CODE_SUBAGENT_MODEL) to a TeamoRouter ID, or delete it.
• Delete CLAUDE_CODE_ENABLE_GATEWAY_MODEL_DISCOVERY and choose models with the variables above.
• If you set these in .claude/settings.local.json under env, make the same changes there. Then restart claude; /status should show https://api.teamorouter.com as the base URL.

Full steps: Claude Code setup.

Codex

Codex reads ~/.codex/config.toml, not OPENAI_BASE_URL.

Before:

model_provider = "openrouter"
model = "~openai/gpt-sol-latest"

[model_providers.openrouter]
name = "openrouter"
base_url = "https://openrouter.ai/api/v1"

[model_providers.openrouter.auth]
command = "sh"
args = ["-c", "echo $OPENROUTER_API_KEY"]

After:

model_provider = "teamorouter"
model = "gpt-6-sol"
model_reasoning_effort = "high"

[model_providers.teamorouter]
name = "TeamoRouter"
base_url = "https://api.teamorouter.com/v1"
env_key = "OPENAI_API_KEY"
wire_api = "responses"

Then set the key with export OPENAI_API_KEY="sk-teamo-xxxxxx". Codex uses the Responses API, which on TeamoRouter serves GPT models only (Claude and Gemini return 400), so pick a GPT model here. Full steps, including Windows: Codex CLI setup.

5. What TeamoRouter doesn't have

6. Check the switch

curl https://api.teamorouter.com/v1/chat/completions \
  -H "Authorization: Bearer sk-teamo-xxxxxx" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-v4-pro","messages":[{"role":"user","content":"Hello"}]}'

• 401: the key is wrong. It should start with sk-teamo-, not sk-or-.
• 402 (400 on the OpenAI format) with "wallet balance is insufficient": the balance is $0. Top up and retry.
• 400 model_not_available: the model ID still has a vendor/ prefix or isn't in our catalog. Check the table above.

More error codes and limits: API integration guide and Rate limits.