API endpoint
All requests go to:
https://www.bettertoken.com/v1
base_url and API Key.
Authentication
Pass your BetterToken API Key in the request:
Anthropic SDK (Claude)
OpenAI SDK (GPT)
cURL
import anthropic
client = anthropic.Anthropic(
base_url="https://www.bettertoken.com/v1",
api_key="YOUR_API_KEY",
)
from openai import OpenAI
client = OpenAI(
base_url="https://www.bettertoken.com/v1",
api_key="YOUR_API_KEY",
)
# Anthropic-compatible endpoint
curl https://www.bettertoken.com/v1/messages \
-H "x-api-key: YOUR_API_KEY" \
-H "anthropic-version: 2023-06-01" \
-H "content-type: application/json" \
-d '{ ... }'
# OpenAI-compatible endpoint
curl https://www.bettertoken.com/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "content-type: application/json" \
-d '{ ... }'
Supported models
Claude series (Anthropic-compatible)
| Model | Description |
|---|
claude-opus-4-5 | Most capable — best for complex reasoning |
claude-sonnet-4-5 | Balanced — recommended for everyday use |
claude-haiku-4-5 | Fastest — best for simple tasks |
GPT series (OpenAI-compatible)
| Model | Description |
|---|
gpt-4o | OpenAI flagship model |
gpt-4o-mini | Lightweight, faster responses |
When using Claude Code or Codex CLI, you do not need to specify a model name in your config. BetterToken routes automatically. You only need to specify a model in tools like Cursor that require explicit model selection.
Environment variable reference
| Variable | Value | Usage |
|---|
ANTHROPIC_BASE_URL | https://www.bettertoken.com/v1 | Anthropic-compatible tools (Claude Code, etc.) |
ANTHROPIC_AUTH_TOKEN | Your API Key | Claude Code authentication |
OPENAI_BASE_URL | https://www.bettertoken.com/v1 | OpenAI SDK-compatible tools |
OPENAI_API_KEY | Your API Key | OpenAI SDK authentication |
API_TIMEOUT_MS | 3000000 | Recommended — prevents long-task timeouts |
CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC | 1 | Claude Code only — reduces background requests |
