Skip to main content
Step Code supports subscription login, API keys, cloud platforms, and OpenAI / Anthropic / Google-compatible endpoints through models.json, including Ollama, vLLM, and LM Studio.

​
Connection methods

When troubleshooting credentials, check CLI --api-key, saved auth.json, provider settings in models.json, and the process environment. If a provider has several credential sources, verify the selected account and endpoint.

​
Subscription login

Choose Step Plan for an account on platform.stepfun.com, or Step Plan Oversea for an account on platform.stepfun.ai. OAuth credentials refresh through the login flow. Use /logout to clear the corresponding credentials. Choosing another channel does not automatically use your subscription quota. Available models and allowances depend on the account. See Step Plan. For sign-in from SSH, server, or container terminals, see Sign in remotely.

​
Step API endpoints

When configuring a Step API key through the environment, set STEP_BASE_URL for the account and billing channel: Bash example for a Step Plan key on stepfun.ai:
Keep the key out of repositories and shared material. Anyone who obtains it can make model requests against your account and incur usage.

​
API key providers

For example, configure Anthropic in Bash or Zsh:
Replace the placeholder with that provider’s key. A Step key cannot be used for another provider. Keys saved through Step sign-in live in ~/.stepcode/auth.json; do not commit this file. See Environment variables for runtime environment variables.

​
Cloud platforms

Deployment names, available regions, and IAM permissions are managed by the cloud platform. Confirm access to the target model before selecting it in Step Code.

​
Custom models (models.json)

Edit ~/.stepcode/models.json, then reopen /model to load changes. This example connects to local Ollama:
Prepare these models in Ollama first. ollama is a nonempty apiKey placeholder. Local services without authentication still require this field to appear in /model; authenticated remote services require real credentials. Model fields include id, name, reasoning, contextWindow, maxTokens, cost, thinkingLevelMap, and samplingParams. id is required. contextWindow defaults to 128,000 and maxTokens to 16,384; override them to match model specifications. samplingParams merges sampling parameters into OpenAI-compatible requests and may override named request fields. Do not add unsupported fields. Setting only baseUrl overrides a built-in provider’s endpoint; models updates matching IDs and adds new ones. Ollama / vLLM / SGLang often need supportsDeveloperRole = false and supportsReasoningEffort = false in compat. See FAQ for model loading and login troubleshooting.

​
Switching models

  • /model opens the picker; Enter saves the startup default. The Recent section at the top lists recently used models, up to 10, with repeats moved to the top; the list is saved globally and shared across sessions.
  • Ctrl+P cycles enabledModels; see Keyboard shortcuts for reverse cycling.
  • Use step --provider <provider> --model <model-ID> on the command line.
  • /effort changes thinking level; available levels and effective behavior depend on the model.
See Configuration files for default model settings.

​
Sign in remotely

In an SSH, server, or container terminal, run step login --no-browser to get an authorization link, then complete sign-in in a browser on this or another device. The command still requires an interactive terminal and does not depend on a local callback port.