> ## Documentation Index
> Fetch the complete documentation index at: https://docs.getinboxzero.com/llms.txt
> Use this file to discover all available pages before exploring further.

# LLM

> Configure your AI provider via environment variables

For self-hosting, configure AI through environment variables in `apps/web/.env`.
If you used `inbox-zero setup`, many of these values are configured automatically.

Start here:

* [Environment Variables](/hosting/environment-variables) (full reference)

<Warning>
  API keys require billing credits on the provider's platform. A ChatGPT Plus or Claude Pro subscription does **not** include API access.
</Warning>

## Providers

Use one of these provider values before the colon in `*_LLMS` entries:

| Provider                                                                        | Value               |
| ------------------------------------------------------------------------------- | ------------------- |
| [OpenAI](https://platform.openai.com/docs/overview)                             | `openai`            |
| [Anthropic](https://docs.anthropic.com/en/docs/welcome)                         | `anthropic`         |
| [Azure OpenAI](https://ai-sdk.dev/providers/ai-sdk-providers/azure)             | `azure`             |
| [Google Gemini (AI Studio)](https://ai.google.dev/gemini-api/docs)              | `google`            |
| [Google Vertex AI](https://ai-sdk.dev/providers/ai-sdk-providers/google-vertex) | `vertex`            |
| [OpenRouter](https://openrouter.ai/docs/quickstart)                             | `openrouter`        |
| [Groq](https://console.groq.com/docs/quickstart)                                | `groq`              |
| [Vercel AI Gateway](https://ai-sdk.dev/providers/ai-sdk-providers/ai-gateway)   | `aigateway`         |
| [AWS Bedrock](https://ai-sdk.dev/providers/ai-sdk-providers/amazon-bedrock)     | `bedrock`           |
| [Ollama](https://ollama.com/)                                                   | `ollama`            |
| OpenAI-compatible (LM Studio, vLLM, LiteLLM, etc.)                              | `openai-compatible` |
| Codex CLI (experimental, self-host only)                                        | `codex-cli`         |
| Claude Code (experimental, self-host only)                                      | `claude-code`       |

## Tiers

For most self-hosted setups, configure the default tier and optionally override cheaper or role-specific tiers:

* `DEFAULT_LLMS` (required): normal AI tasks.
* `ECONOMY_LLMS` (optional): lower-cost model for high-volume tasks. If unset, it falls back to default.
* `CHAT_LLMS` (optional): assistant chat. If unset, it falls back to default.
* `DRAFT_LLMS` (optional): reply drafting. If unset, it falls back to default.
* `NANO_LLMS` (optional): lightweight classification and extraction. If unset, it falls back to economy/default behavior.

Each value is an ordered comma-separated list in `provider:model` format. The first valid entry is the primary model. Later valid entries are ordered fallbacks. Model names can contain colons; only the first colon separates the provider from the model.

Minimal example:

```env theme={null}
DEFAULT_LLMS=openai:gpt-5.4-mini
ECONOMY_LLMS=openai:gpt-5.4-nano
LLM_API_KEY=sk-...
```

Provider-specific keys (for example `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`) also work. See [Environment Variables](/hosting/environment-variables) for the full list.

## App Settings

The app also has **Settings → AI** for per-user keys/models, but self-hosted deployments usually keep configuration at the environment-variable level.

## Sensitive Data Protection

Self-hosted deployments can choose how LLM requests handle sensitive data matches before they are sent to an AI provider. The current scanner targets likely credentials/tokens and payment-card-like numbers; it is not a full DLP or PHI classifier.

```env theme={null}
SENSITIVE_DATA_POLICY_DEFAULT=ALLOW # ALLOW, REDACT, or BLOCK
NEXT_PUBLIC_SENSITIVE_DATA_POLICY_LOCKED=false
```

`ALLOW` preserves the default behavior. `REDACT` replaces matched values before the LLM request. `BLOCK` stops the request when a match is found. Leave `NEXT_PUBLIC_SENSITIVE_DATA_POLICY_LOCKED=false` to let users choose per account in Settings, or set it to `true` to enforce the deployment default for all accounts and hide the setting from the UI.

## Provider-specific details

* `openai-compatible` also requires `OPENAI_COMPATIBLE_BASE_URL`.
* `ollama` can use `OLLAMA_BASE_URL` and `OLLAMA_MODEL`.

## Fallbacks

Add fallbacks by adding more entries to the same role list:

```env theme={null}
DEFAULT_LLMS=openrouter:anthropic/claude-sonnet-4.6,openai:gpt-5.4-mini
ECONOMY_LLMS=openrouter:google/gemini-2.5-flash,openai:gpt-5.4-nano
CHAT_LLMS=openrouter:anthropic/claude-haiku-4.5,openai:gpt-5.4-mini
```

Unsupported providers, entries without configured credentials, entries without models, and duplicates are skipped with warnings.

### Legacy variables

The old `DEFAULT_LLM_PROVIDER` / `DEFAULT_LLM_MODEL`, role-specific `*_LLM_PROVIDER` / `*_LLM_MODEL`, and `*_LLM_FALLBACKS` variables are deprecated but still supported for existing deployments. At startup, they are converted into the corresponding `*_LLMS` value. New deployments and CLI-generated env files should use `*_LLMS`.

### CLI LLM providers

`codex-cli` and `claude-code` are experimental self-host options. They use
third-party community AI SDK provider packages that spawn local CLI tools, so
they are disabled unless `CLI_LLM_ENABLED=true`.

Use them only on trusted self-hosted deployments. Review the provider package
source, pin exact package versions, and make sure you comply with the relevant
OpenAI or Anthropic terms for your authentication method.

Codex example:

```bash theme={null}
cd apps/web
pnpm add ai-sdk-provider-codex-cli@1.1.0
pnpm add -g @openai/codex
codex login
```

```env theme={null}
CLI_LLM_ENABLED=true
DEFAULT_LLMS=codex-cli:gpt-5.3-codex
CODEX_CLI_ALLOW_NPX=false
```

Claude Code example:

```bash theme={null}
cd apps/web
pnpm add ai-sdk-provider-claude-code@3.1.0
claude auth login
```

```env theme={null}
CLI_LLM_ENABLED=true
DEFAULT_LLMS=claude-code:sonnet
```