How to Use OpenRouter model search in Grok Build
Returns matching models from OpenRouter's catalog for comparison. Use it to build a shortlist, then inspect individual model specifications and serving endpoints.
Prompts to try with OpenRouter model search in Grok Build
Copy an example into Grok Build, or ask in your own words.
Use OpenRouter model search to find Claude models, then add a shortlist of returned model names and context windows to our evaluation notes.
Use OpenRouter model search to look this up and give me the result as a table. Save the result in ./assets.
Pull the latest numbers from OpenRouter model search and tell me what changed since last week.
Check OpenRouter model search and summarize what you find in three bullet points.
Pull the newest data from OpenRouter model search and give me the result as a table.
How to set up OpenRouter model search in Grok Build
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the server
Add --scope project to add Spiral to the current repository only. No header and no client ID or secret: Grok registers itself with Spiral.
grok mcp add --transport http spiral https://api.spiral.ai/mcp
- Sign in
Start grok and run /mcps. Select Spiral and press i. Your browser opens: sign in to Spiral, pick the organization and click Authorize. No API key.
- Check it
grok mcp doctor spiral
Use the Spiral skill
- Send Grok Build this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run OpenRouter model search
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "alexandria", "model": "openrouter-ai/model-marketplace/search_models", "input": { "q": "<q>" } }'
Then paste one of the prompts above: Grok Build finds OpenRouter model search on Spiral, shows the price and runs it.
OpenRouter model search providers and prices
Spiral runs OpenRouter model search through Alexandria. Open a card for today's price, with every fee included.
Settings you can ask for
Ask Grok Build for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Zdr | on or off | Only models with a zero-data-retention endpoint. | |
| Sort | most-popular, newest, top-weekly, pricing-low-to-high, pricing-high-to-low, context-high-to-low, throughput-high-to-low, latency-low-to-high and 4 more | Server-side order. | |
| Limit | 1 to 100 | 25 | limit |
| Offset | a number from 0 | 0 | Records to skip; use |
| Region | eu or us | Only models with endpoints in this data region. | |
| Category | programming, roleplay, marketing, marketing/seo, technology, science, translation, legal and 4 more | Use-case category filter. Cannot be combined with | |
| Max price | a number from 0 | Maximum prompt price in USD per million tokens (0 lists free models). | |
| Min price | a number from 0 | Minimum prompt price in USD per million tokens. | |
| Min context | a number from 1 | Minimum context length in tokens. |
Every API field of OpenRouter · Search models is on its tool page: open its card above.
What can you do with OpenRouter model search in Grok Build?
Build a model selection report around the requirements in your project. An agent can compare OpenRouter catalog evidence with your document sizes and output needs, then identify candidates for testing.
Choose a model for document summaries
Find candidates, compare their context windows and output capabilities, inspect serving limits, and review usage statistics before selecting models for a document-summary evaluation.
Prepare a model migration review
Search for replacements, review the author's catalog, compare model details and pricing data, and summarize the provider policies relevant to the shortlist.
Maintain an internal model comparison
Refresh a shortlist with published specifications, pricing data, usage statistics, and serving endpoint measurements. Use the documentation to keep metric definitions consistent.
Prepare a provider policy review
Read OpenRouter's policy documentation, review its provider directory, find candidate models, and inspect the specific serving endpoints before preparing questions for a policy review.
Write a model selection brief
Combine catalog search, an author's listed models, pricing data, usage statistics, provider policies, and documentation into a source-based brief for a product team.
More on OpenRouter model search and Grok Build
Overview
OpenRouter model search in other agents
Questions about OpenRouter model search in Grok Build
Can Grok Build use OpenRouter model search?
Yes. Add Spiral to Grok Build as an MCP server, and Grok Build can find and run OpenRouter model search like any built-in tool.
How much does OpenRouter model search cost in Grok Build?
$0.00873 per call. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Grok Build plan is separate.
Do I need an Alexandria account to use OpenRouter model search in Grok Build?
No. Spiral runs OpenRouter model search on its own Alexandria account. You sign in to Spiral once and never paste a key.
How long does it take to set up OpenRouter model search in Grok Build?
About two minutes: add the server, then sign in, then check it.
Does Grok Build wait while OpenRouter model search runs?
OpenRouter model search runs in the background. Grok Build starts the run, checks on it and gives you the json when it's ready.
Does OpenRouter's default model list cover every output type?
The documented default list covers text generation. OpenRouter also documents a broader listing across all modalities, including image, video, text, and speech.
How quickly does OpenRouter update model information?
OpenRouter says it publishes model information as soon as it confirms it. That statement does not establish a fixed hourly or daily refresh schedule.
What time period do performance measurements cover?
The routing guide describes rolling five-minute percentiles, while the endpoint response schema defines latency percentiles over 30 minutes. These are different measurement windows, so comparisons need to use the same window.
Does the context window apply only to the input?
Input and output share the context window. A long input therefore reduces the space available for the response, even when the model lists a larger maximum output ceiling.
Are OpenRouter's coding and agentic benchmark fields still populated?
The current schema says Artificial Analysis retired its Coding Index and Agentic Index. Both fields are null for every model.
Start using OpenRouter model search in Grok Build
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Grok Build:
set up https://spiral.ai/SKILL.md


