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

> ## Agent Instructions
> To act on a user's behalf in Guild, use the Guild REST API at https://api.guild.ai/v1.
> The OpenAPI spec is unauthenticated: https://docs.guild.ai/openapi.json (or openapi.yaml).
> Authenticate with an account API key, sent as HTTP Basic (key id as username, secret as password) or as `Authorization: Bearer <api_key_id>:<api_key_secret>`. See https://docs.guild.ai/api-reference/introduction#authentication.
> To run one agent on demand without account-wide access, use an API trigger instead: https://docs.guild.ai/platform/api-triggers.
> To search these docs from an MCP client, connect to https://docs.guild.ai/mcp. The full page index is https://docs.guild.ai/llms.txt.

# Reach every model Guild supports with one OpenRouter key

> Connect OpenRouter to Guild with one API key, and run Claude, GPT, Gemini, Llama, DeepSeek, Qwen, Kimi, and GLM models from your agents, under policy.

Integrations · OpenRouter

Guild agents call models from every publisher Guild supports through one OpenRouter API key. Guild holds the key server-side, checks your model policies on every call, and OpenRouter bills the calls to your account.

> **[Book a demo](https://guild-ai.cal.com/gtm/meeting-with-guild.ai)**
>
> See how teams govern agents that use OpenRouter.

> **[Start free](https://app.guild.ai)**
>
> Connect OpenRouter and scope your first agent.

## One key, eight publishers

OpenRouter routes to Anthropic, OpenAI, Gemini, Meta, DeepSeek, Alibaba, Moonshot AI, and Z.ai. Guild refreshes its list from OpenRouter's public catalog every day. You write canonical names such as `claude-sonnet-4-6`, and Guild adds the namespace and OpenRouter's spelling, such as `anthropic/claude-sonnet-4.6`.

## How it works

**1. Create an OpenRouter key**

In your OpenRouter account, create an API key.

**2. Add the credential**

In **Access & setup > Models & providers**, click **Add API key** and choose **OpenRouter**. Paste the key and check every publisher this credential should reach under **Models**.

**3. Scope it with policies**

Add model policies that decide which workspaces and agents may call which models, then call them from `task.llm` by canonical name.

## Every OpenRouter call goes through your policies

Agents never hold the key. Guild stores the credential server-side and shows only a masked value, so you can rotate it without touching agent code or the [model policies](/platform/llm-settings#model-policies) bound to it. Policies are an allowlist: a workspace, agent, or installed agent can call a model through it only if a rule allows it. Set a [daily token limit](/platform/llm-settings#daily-token-limit) to cap spend, and see each call by workspace, agent, and user in [Usage](/insights/usage).

A raw API key in an agent's environment reaches every model the key can. A policy reaches only the models you name.

## About OpenRouter on Guild

| | |
| - | - |
| **Inference provider** in the app | OpenRouter |
| Authentication | An OpenRouter API key |
| Model IDs | Mapped from canonical names for you |

## Publishers on OpenRouter

| Publisher | OpenRouter namespace | Page |
| - | - | - |
| Anthropic | `anthropic/` | [Anthropic](/platform/models/anthropic) |
| OpenAI | `openai/` | [OpenAI](/platform/models/openai) |
| Gemini | `google/` | [Gemini](/platform/models/gemini) |
| Meta | `meta/` for Muse, `meta-llama/` for Llama | [Meta](/platform/models/meta) |
| DeepSeek | `deepseek/` | [DeepSeek](/platform/models/deepseek) |
| Alibaba | `qwen/` | [Qwen](/platform/models/qwen) |
| Moonshot AI | `moonshotai/` | [Moonshot AI](/platform/models/moonshot) |
| Z.ai | `z-ai/` | [Z.ai](/platform/models/zai) |

Guild refreshes its list from OpenRouter's public model catalog every day, keeping the text models in these namespaces. Variants with a suffix after a colon, such as `:free`, aren't listed. To see what's available, browse [OpenRouter's models](https://openrouter.ai/models) and drop the namespace: `qwen/qwen3-coder` becomes `qwen3-coder`.

## Connect OpenRouter

**1. Create a key**

In your OpenRouter account, create an API key.

**2. Start the credential**

In Guild, click **Access & setup** in the left nav, then **Models & providers**. Click **Add API key** and choose **OpenRouter** from the menu.

**3. Name it and paste the key**

Give the credential a **Name**, such as `Production`, and paste the key into **API key**.

**4. Choose publishers**

Under **Models**, check each publisher this credential should reach. Keep or change the default model next to each one.

**5. Save**

Click **Add Key**.

One OpenRouter credential can reach every publisher at once. Check them all to use a single key for everything.

## Model IDs

Write canonical model names, such as `claude-sonnet-4-6` or `deepseek-r1`, in policies and `llmPreferences`, without the OpenRouter namespace. Guild adds the namespace for you. OpenRouter spells Claude versions with dots, so Guild maps `claude-sonnet-4-6` to `anthropic/claude-sonnet-4.6`.

## Pricing

OpenRouter bills your OpenRouter account at its own rates, and those rates change as OpenRouter and the model publishers update them. For current prices, check each model's page on [OpenRouter's models list](https://openrouter.ai/models), and see [OpenRouter pricing](https://openrouter.ai/pricing) for how OpenRouter charges.

Guild's [Usage](/insights/usage) pages estimate spend from Guild's own price list, which can lag behind OpenRouter's. For a model Guild hasn't priced yet, the estimate uses a fallback rate. Your OpenRouter account is the source of truth for what you paid.

## Frequently asked questions

> **Can one OpenRouter key cover every publisher?**
>
> Yes. Check every publisher under **Models** when you add the credential, and one key reaches all of them.

> **Do Guild's usage numbers match my OpenRouter bill?**
>
> Not exactly. Guild estimates spend from its own price list, which can lag behind OpenRouter's rates, and uses a fallback rate for models it hasn't priced. Your OpenRouter account is the source of truth for what you paid.

> **Can I use free variants such as :free?**
>
> No. Variants with a suffix after a colon aren't listed.

> **Do I include the OpenRouter namespace in model names?**
>
> No. Write canonical names, such as `deepseek-r1`, in policies and `llmPreferences`. Guild adds the namespace for you.

## Put a governed agent on OpenRouter this week

> **[Book a demo](https://guild-ai.cal.com/gtm/meeting-with-guild.ai)**
>
> See how teams govern agents that use OpenRouter.

> **[Start free](https://app.guild.ai)**
>
> Connect OpenRouter and scope your first agent.

## Related

**Also runs on Guild:** [Anthropic](/integrations/anthropic) · [Gemini](/integrations/gemini) · [AWS Bedrock](/integrations/aws-bedrock) · [Fireworks AI](/integrations/fireworks)

**Read:** Docs: [Models & providers](/platform/llm-settings) · [LLMs](/guide/llms) · [Usage](/insights/usage)