One API. Leading AI models.

1. Create an API key

Open the console, go to API Keys, create a key with the model group you need, and store it securely.

Test your connection with a compatible chat completion request:

curl https://modviax.com/v1/chat/completions \
  -H "Authorization: Bearer $MODVIAX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-5.6-sol","messages":[{"role":"user","content":"Hello"}]}'

2. Configure Claude Code

Install Claude Code, then add the following environment variables to your shell or settings file:

ANTHROPIC_AUTH_TOKEN=sk-your-api-key
ANTHROPIC_BASE_URL=https://modviax.com
ANTHROPIC_MODEL=claude-opus-5

Update Claude Code with claude update when needed.

3. Configure Codex

Install the Codex CLI and place your key in ~/.codex/auth.json. Set the compatible provider in config.toml:

model_provider = "custom"
model = "gpt-5.6-sol"
model_reasoning_effort = "xhigh"
disable_response_storage = true

[model_providers.custom]
name = "custom"
wire_api = "responses"
requires_openai_auth = true
base_url = "https://modviax.com/v1"

4. Use CC-Switch

CC-Switch makes it easy to manage Claude Code and Codex profiles. Add a profile, enter your API key and base URL, select the model, then save and activate it.

5. Connect OpenCode

Add a provider in ~/.config/opencode/opencode.json:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {"anthropic": {"npm": "@ai-sdk/anthropic", "name": "modviax", "options": {"baseURL": "https://modviax.com/v1", "apiKey": "sk-your-api-key"}}}
}