Provider Setup Guide

This guide covers detailed setup instructions for every provider supported by Claudex, including API key providers, OAuth subscriptions, cloud platforms, and local models.

API Key Providers

Anthropic

1. Sign up at console.anthropic.com
2. Navigate to API Keys and create a new key (format: sk-ant-*)

[[profiles]]
name = "anthropic"
provider_type = "DirectAnthropic"
base_url = "https://api.anthropic.com"
api_key = "sk-ant-..."
default_model = "claude-sonnet-4-20250514"

claudex profile test anthropic

MiniMax

1. Sign up at platform.minimaxi.com
2. Navigate to API Keys in the console

[[profiles]]
name = "minimax"
provider_type = "DirectAnthropic"
base_url = "https://api.minimax.io/anthropic"
api_key = "..."
default_model = "claude-sonnet-4-20250514"

claudex profile test minimax

OpenRouter

1. Sign up at openrouter.ai
2. Get your API key at openrouter.ai/keys

[[profiles]]
name = "openrouter"
provider_type = "OpenAICompatible"
base_url = "https://openrouter.ai/api/v1"
api_key = "sk-or-..."
default_model = "anthropic/claude-sonnet-4"

claudex profile test openrouter

Tip

OpenRouter supports hundreds of models from different providers. Use the anthropic/ prefix for Claude models.

Grok (xAI)

1. Sign up at console.x.ai
2. Create an API key in the console

[[profiles]]
name = "grok"
provider_type = "OpenAICompatible"
base_url = "https://api.x.ai/v1"
api_key = "xai-..."
default_model = "grok-3-beta"

claudex profile test grok

OpenAI

1. Sign up at platform.openai.com
2. Create an API key at platform.openai.com/api-keys

[[profiles]]
name = "chatgpt"
provider_type = "OpenAICompatible"
base_url = "https://api.openai.com/v1"
api_key = "sk-..."
default_model = "gpt-4o"

claudex profile test chatgpt

DeepSeek

1. Sign up at platform.deepseek.com
2. Create an API key in the console

[[profiles]]
name = "deepseek"
provider_type = "OpenAICompatible"
base_url = "https://api.deepseek.com"
api_key = "sk-..."
default_model = "deepseek-chat"

claudex profile test deepseek

Kimi / Moonshot

1. Sign up at platform.moonshot.cn
2. Create an API key in the console

[[profiles]]
name = "kimi"
provider_type = "OpenAICompatible"
base_url = "https://api.moonshot.ai/v1"
api_key = "sk-..."
default_model = "kimi-k2-0905-preview"

claudex profile test kimi

GLM / Zhipu

1. Sign up at open.bigmodel.cn
2. Create an API key in the console

[[profiles]]
name = "glm"
provider_type = "OpenAICompatible"
base_url = "https://api.z.ai/api/paas/v4"
api_key = "..."
default_model = "glm-4.6"

claudex profile test glm

Groq

1. Sign up at console.groq.com
2. Create an API key in Settings > API Keys

[[profiles]]
name = "groq"
provider_type = "OpenAICompatible"
base_url = "https://api.groq.com/openai/v1"
api_key = "gsk_..."
default_model = "llama-3.3-70b-versatile"

claudex profile test groq

Tip

Groq provides extremely fast inference for open-source models. Popular models include llama-3.3-70b-versatile, mixtral-8x7b-32768, and gemma2-9b-it.

Mistral AI

1. Sign up at console.mistral.ai
2. Create an API key in the console

[[profiles]]
name = "mistral"
provider_type = "OpenAICompatible"
base_url = "https://api.mistral.ai/v1"
api_key = "..."
default_model = "mistral-large-latest"

claudex profile test mistral

Together AI

1. Sign up at api.together.ai
2. Create an API key in the dashboard

[[profiles]]
name = "together"
provider_type = "OpenAICompatible"
base_url = "https://api.together.xyz/v1"
api_key = "..."
default_model = "meta-llama/Llama-3.3-70B-Instruct-Turbo"

claudex profile test together

Perplexity

1. Sign up at perplexity.ai
2. Create an API key at perplexity.ai/settings/api

[[profiles]]
name = "perplexity"
provider_type = "OpenAICompatible"
base_url = "https://api.perplexity.ai"
api_key = "pplx-..."
default_model = "sonar-pro"

claudex profile test perplexity

Note

Perplexity models have built-in web search capabilities. Useful for research-oriented tasks.

Cerebras

1. Sign up at cloud.cerebras.ai
2. Create an API key in the dashboard

[[profiles]]
name = "cerebras"
provider_type = "OpenAICompatible"
base_url = "https://api.cerebras.ai/v1"
api_key = "..."
default_model = "llama-3.3-70b"

claudex profile test cerebras

Cloud Platforms

Azure OpenAI

1. Create an Azure OpenAI resource in the Azure Portal
2. Deploy a model and note the resource name and deployment name
3. Get your API key from Keys and Endpoint

[[profiles]]
name = "azure-openai"
provider_type = "OpenAICompatible"
base_url = "https://YOUR_RESOURCE.openai.azure.com/openai/deployments/YOUR_DEPLOYMENT"
api_key = "YOUR_AZURE_KEY"
default_model = "gpt-4o"

[profiles.query_params]
api-version = "2024-12-01-preview"

claudex profile test azure-openai

Note

Azure OpenAI is auto-detected by base_url containing openai.azure.com. Authentication uses the api-key header instead of Authorization: Bearer. The api-version query parameter is required.

Google Vertex AI

1. Enable the Vertex AI API in your GCP project
2. Generate an access token with gcloud auth print-access-token

[[profiles]]
name = "vertex-ai"
provider_type = "DirectAnthropic"
base_url = "https://us-east5-aiplatform.googleapis.com/v1/projects/YOUR_PROJECT/locations/us-east5/publishers/anthropic/models"
api_key = "YOUR_GCLOUD_TOKEN"
default_model = "claude-sonnet-4@20250514"

claudex profile test vertex-ai

Caution

The gcloud auth print-access-token output expires after ~1 hour. For production use, consider a service account with token refresh automation.

AWS Bedrock (via LiteLLM)

AWS Bedrock is supported through a LiteLLM proxy:

1. Install LiteLLM: pip install litellm
2. Start the proxy:

litellm --model bedrock/anthropic.claude-sonnet-4-20250514-v2:0

3. Configure Claudex:

[[profiles]]
name = "bedrock"
provider_type = "OpenAICompatible"
base_url = "http://localhost:4000/v1"
api_key = "sk-litellm"
default_model = "bedrock/anthropic.claude-sonnet-4-20250514-v2:0"

OAuth Subscriptions

OAuth authentication lets you use existing provider subscriptions (ChatGPT Plus, Claude Max, etc.) instead of separate API keys.

Claude Max / Pro

Use your existing Claude subscription through Claude Code’s native OAuth session.

Prerequisites: Claude Code installed and logged in (claude command works normally)

[[profiles]]
name = "claude-max"
provider_type = "DirectAnthropic"
base_url = "https://api.claude.ai"
default_model = "claude-sonnet-4-20250514"
auth_type = "oauth"
oauth_provider = "claude"

[profiles.models]
haiku = "claude-haiku-4-20250514"
sonnet = "claude-sonnet-4-20250514"
opus = "claude-opus-4-20250514"

Note

Claude OAuth profiles use a special gateway auth mode. Claudex skips the proxy and lets Claude Code use its own OAuth directly. No separate claudex auth login step is needed.

ChatGPT Plus / Codex CLI

Use your ChatGPT Plus or Pro subscription via the Codex CLI token.

Prerequisites: Install Codex CLI and authenticate:

# Install Codex CLI
npm install -g @openai/codex

# Authenticate (opens browser)
codex auth

Then configure Claudex:

[[profiles]]
name = "codex-sub"
provider_type = "OpenAIResponses"
base_url = "https://chatgpt.com/backend-api/codex"
default_model = "gpt-5.3-codex"
auth_type = "oauth"
oauth_provider = "openai"

[profiles.models]
haiku = "codex-mini-latest"
sonnet = "gpt-5.3-codex"
opus = "gpt-5.3-codex"

# Read token from Codex CLI
claudex auth login openai --profile codex-sub

# Verify
claudex auth status

# Run
claudex run codex-sub

Tip

Supported models on the Codex endpoint: gpt-5.3-codex, gpt-5.1-codex, codex-mini-latest. Note that gpt-4o is NOT supported on the chatgpt.com Codex endpoint. strip_params = "auto" will automatically detect this endpoint and strip temperature/top_p/top_k.

Google Gemini

Use your Google account via Gemini CLI credentials.

[[profiles]]
name = "gemini-sub"
provider_type = "OpenAICompatible"
base_url = "https://generativelanguage.googleapis.com/v1beta/openai"
default_model = "gemini-2.5-pro"
auth_type = "oauth"
oauth_provider = "google"

[profiles.models]
haiku = "gemini-2.0-flash"
sonnet = "gemini-2.5-pro"
opus = "gemini-2.5-pro"

# Login reads from Gemini CLI credentials
claudex auth login google --profile gemini-sub

# Verify
claudex auth status

Qwen / Tongyi Qianwen

Use your Qwen account via OAuth Device Code flow.

[[profiles]]
name = "qwen-oauth"
provider_type = "OpenAICompatible"
base_url = "https://chat.qwen.ai/api"
default_model = "qwen3-235b-a22b"
auth_type = "oauth"
oauth_provider = "qwen"

# Start device code flow
claudex auth login qwen --profile qwen-oauth

# Verify
claudex auth status

Kimi / Moonshot (OAuth)

Use your Kimi account via Kimi CLI credentials.

[[profiles]]
name = "kimi-oauth"
provider_type = "OpenAICompatible"
base_url = "https://api.moonshot.cn/v1"
default_model = "moonshot-v1-128k"
auth_type = "oauth"
oauth_provider = "kimi"

# Login reads from Kimi CLI credentials
claudex auth login kimi --profile kimi-oauth

# Verify
claudex auth status

GitHub Copilot

Use your GitHub Copilot subscription via OAuth Device Code flow.

[[profiles]]
name = "copilot"
provider_type = "OpenAICompatible"
base_url = "https://api.githubcopilot.com"
default_model = "gpt-4o"
auth_type = "oauth"
oauth_provider = "github"

# Start device code flow (opens browser for GitHub login)
claudex auth login github --profile copilot

# Verify
claudex auth status

GitLab Duo

Use your GitLab Duo subscription via a Personal Access Token.

[[profiles]]
name = "gitlab-duo"
provider_type = "OpenAICompatible"
base_url = "https://gitlab.com/api/v4/ai/llm/proxy"
default_model = "claude-sonnet-4-20250514"
auth_type = "oauth"
oauth_provider = "gitlab"

# Set your GitLab token as an environment variable
export GITLAB_TOKEN=glpat-...

# Login
claudex auth login gitlab --profile gitlab-duo

# Verify
claudex auth status

Note

GitLab Duo requires a GITLAB_TOKEN environment variable with a Personal Access Token that has the ai_features scope. For self-managed GitLab instances, update base_url accordingly.

Local Models

Ollama

1. Install Ollama: ollama.com
2. Pull a model:

ollama pull qwen2.5:72b

3. Configure Claudex:

[[profiles]]
name = "local-qwen"
provider_type = "OpenAICompatible"
base_url = "http://localhost:11434/v1"
api_key = ""
default_model = "qwen2.5:72b"

claudex profile test local-qwen

Note

Ollama does not require an API key. Leave api_key empty or omit it.

vLLM

1. Install and start vLLM:

pip install vllm
vllm serve meta-llama/Llama-3.3-70B-Instruct --port 8000

2. Configure Claudex:

[[profiles]]
name = "local-llama"
provider_type = "OpenAICompatible"
base_url = "http://localhost:8000/v1"
api_key = ""
default_model = "meta-llama/Llama-3.3-70B-Instruct"

LM Studio

1. Download LM Studio from lmstudio.ai
2. Load a model and start the local server (default port: 1234)

[[profiles]]
name = "lm-studio"
provider_type = "OpenAICompatible"
base_url = "http://localhost:1234/v1"
api_key = "lm-studio"
default_model = "local-model"

Verifying Your Setup

After configuring any provider, verify connectivity:

# Test a specific profile
claudex profile test <profile-name>

# Test all profiles
claudex profile test all

# List all configured profiles
claudex profile list