stella models

List configured providers and available models, and manage the live model catalog.

List the configured providers and their available models, and manage the live model catalog. For each provider, Stella shows its key status, default model, and base URL.

Synopsis

stella models [global flags]
stella models refresh [--force]
stella models list [--provider <id>]

stella models enumerates the built-in providers and reports, for each one, its display name, key status, default model, and base URL, followed by a status line for the local model catalog. Any custom providers you define in settings.json are listed too, under a separate Config-defined providers heading with their wire dialect. It is the quickest way to confirm which providers are ready to use before you pin one with --model.

The key status reflects only whether the provider's environment variable (primary or an alias) is set, or a non-empty api_key is present in settings.json. A key stored only in ~/.config/stella/credentials.toml (or supplied at the interactive prompt) resolves fine at run time but does not light up this column.

The same listing is available from inside an interactive session via the /models slash command.

Subcommands — the live model catalog

The catalog behind model validation and pricing syncs from models.dev into local SQLite, with versioned pricing model cards and strict slug validation (a mistyped model gets a did-you-mean, not a silent request). See Models & the live catalog for the full design.

`stella models refresh [--force]`

Sync the catalog now. Incremental by default (ETag-aware); --force re-downloads everything. Without an explicit refresh, the catalog auto-resyncs when it is older than 24 hours (disable with STELLA_CATALOG_AUTO_REFRESH=0).

stella models refresh
stella models refresh --force

`stella models list [--provider <id>]`

List the models in the local catalog — slug, display name, context window, and current pricing — optionally filtered to one provider.

stella models list
stella models list --provider anthropic

Provider credentials are resolved through the credential chain: --api-key flag, then the provider env var (and aliases), then settings.json, then ~/.config/stella/credentials.toml, and finally an interactive prompt on a TTY.

Flags

stella models takes the global flags, but the listing itself depends only on your settings.json and environment variables — --model, --base-url, and --api-key do not change its output.

Example

stella models

Representative output (illustrative — one line per provider, in auto-detect preference order):

Stella — Available Providers & Models

  ✓ configured zai/glm-5.2  Z.ai (GLM 5.2)  [https://api.z.ai/api/paas/v4]
  ✓ configured anthropic/claude-fable-5  Anthropic (Claude)  [https://api.anthropic.com]
  ✗ no key openai/gpt-5.5  OpenAI (GPT)  [https://api.openai.com/v1]
  ✗ no key xai/grok-4  xAI (Grok)  [https://api.x.ai/v1]
  ✗ no key deepseek/deepseek-chat  DeepSeek  [https://api.deepseek.com/v1]
  ✗ no key gemini/gemini-3-pro  Google Gemini  [https://generativelanguage.googleapis.com/v1beta]
  ✗ no key openrouter/openrouter/auto  OpenRouter  [https://openrouter.ai/api/v1]
  ✗ no key vertex/gemini-3-pro  Google Vertex AI  [https://aiplatform.googleapis.com]
  ✗ no key bedrock/us.anthropic.claude-sonnet-4-5-20250929-v1:0  Amazon Bedrock  [https://bedrock-runtime.<AWS_REGION>.amazonaws.com]

  Config-defined providers (settings.json):
  ✓ configured myproxy/<model>  My proxy  [https://llm.internal.example/v1] (OpenAI-compatible)

  Use --model provider/model_id to pin a specific model.
  Example: stella --model zai/glm-5.2 run 'fix the failing test'
  Local endpoints (Ollama, vLLM, LM Studio): stella --model local/<model> --base-url http://localhost:11434/v1

  Model catalog: 1204 models / 118 aliases — master list refreshed 2026-07-18 09:14 UTC (`stella models list` to browse, `stella models refresh` to update).

Each row is {key status} {id}/{default model} {display name} [{base url}]. The vertex and bedrock base URLs are display anchors — the real endpoint is built per request from your project/location (VERTEX_PROJECT_ID, VERTEX_LOCATION) or AWS_REGION. Config-defined providers from settings.json follow under their own heading with the wire dialect appended. The Config-defined providers section only appears when settings.json defines one, and the catalog line reads seed only until your first stella models refresh.

local is not a provider row — it is surfaced only as the footer hint that reminds you to point at a local endpoint yourself, e.g. stella --model local/<model> --base-url http://localhost:11434/v1.

The values above are for illustration. Your actual key status depends on which credentials resolve in your environment, and default models and base URLs can be overridden per provider via settings.json.