How to Use Hugging Face model search in Hermes
Find model repositories on Hugging Face and compare their metadata, downloads, and likes. Use it to shortlist models for a task before examining a particular repository.
Prompts to try with Hugging Face model search in Hermes
Copy an example into Hermes, or ask in your own words.
Each Monday, use Hugging Face model search to list the ten most downloaded sentence-transformers models for sentence similarity and report changes in the shortlist since the previous check.
Every Monday at 9am: use Hugging Face model search to look this up and give me the result as a table.
Pull the latest numbers from Hugging Face model search and tell me what changed since last week.
Check Hugging Face model search and summarize what you find in three bullet points.
Pull the newest data from Hugging Face model search and give me the result as a table.
How to set up Hugging Face model search in Hermes
Pick the way that suits you. Each takes about two minutes.
Use the Spiral skill
- Send Hermes 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 Spiral's MCP server
- Add the server and sign in
Hermes opens your browser: sign in to Spiral and click Authorize. No API key. When Hermes asks Enable all tools? \[Y/n/select], answer Y: n cancels without saving the server, and select hides the tools you leave out.
hermes mcp add spiral \ --url https://api.spiral.ai/mcp \ --auth oauth \ --connect-timeout 300
- Check it
In a chat that was already open, run /reload-mcp.
hermes mcp list hermes mcp test spiral
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 Hugging Face 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": "huggingface-co/hub/models", "input": { "sort": "downloads" } }'
Then paste one of the prompts above: Hermes finds Hugging Face model search on Spiral, shows the price and runs it.
Hugging Face model search providers and prices
Spiral runs Hugging Face model search through Alexandria. Open a card for today's price, with every fee included.
Settings you can ask for
Ask Hermes for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Sort | downloads, likes, trending, last_modified, or created | downloads | Sort key; always descending. |
| Limit | 1 to 100 | 20 | limit |
Every API field of Hugging Face · Models is on its tool page: open its card above.
What can you do with Hugging Face model search in Hermes?
Keep a recurring review of selected Hugging Face repositories ready for your team. A scheduled agent can compare returned metadata and card content with previous checks and summarize changes.
Prepare a repository license review
Shortlist models and datasets, collect their declared license metadata, and read their cards for the accompanying usage statements. Produce a review table with links to the original repositories.
Plan a model evaluation
Find candidate models and evaluation datasets, inspect their repository metadata, and extract documented tasks and limitations from their cards. Turn the findings into a proposed evaluation shortlist.
Build an internal ML catalog
Discover repositories from selected public organizations, inspect known model and dataset records, and add card summaries to a searchable internal catalog.
Check documentation before adoption
Find relevant model and dataset repositories, inspect their metadata, and read their cards. Build a checklist of documented uses, citations, and limitations that still need review.
Choose a model for a project
Search for models suited to the project's task, inspect each candidate's task and library information, and summarize the limitations documented in its card.
Compare repository interest
Compare downloads and likes across shortlisted repositories. Keep model and dataset download comparisons separate because Hugging Face counts them differently.
More on Hugging Face model search and Hermes
Overview
Hugging Face model search in other agents
Questions about Hugging Face model search in Hermes
Can Hermes use Hugging Face model search?
Yes. Add Spiral to Hermes as an MCP server, and Hermes can find and run Hugging Face model search like any built-in tool.
How much does Hugging Face model search cost in Hermes?
$0.00873 per call. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Hermes plan is separate.
Do I need an Alexandria account to use Hugging Face model search in Hermes?
No. Spiral runs Hugging Face 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 Hugging Face model search in Hermes?
About two minutes: add the server and sign in, then check it.
Does Hermes wait while Hugging Face model search runs?
Hugging Face model search runs in the background. Hermes starts the run, checks on it and gives you the json when it's ready.
Does this cover models and datasets outside Hugging Face?
The data comes from repositories on Hugging Face Hub. A model or dataset hosted only elsewhere is outside this repository coverage.
How complete are repository records?
Many Hub metadata attributes are optional, and search listings contain fewer attributes than individual repository responses. Model cards are recommended to document training, evaluations, and limitations, but that recommendation does not establish that every card includes them.
What period do download counts cover?
The standard Hub downloads attribute covers the last 30 days for models and datasets. Cumulative downloads are defined separately.
Are download counts counts of people?
Model downloads count requests to designated files, including GET and HEAD requests. Dataset downloads combine files fetched from one repository within a five-minute window for the same IP-based user, so the two totals use different counting rules.
Where do license and language labels come from?
Hugging Face uses repository card metadata for these labels. Dataset cards can declare license, language, and size, while model cards also carry publisher-supplied metadata.
Start using Hugging Face model search in Hermes
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Hermes:
set up https://spiral.ai/SKILL.md


