Skip to main content

GET /api/v1/models

The public list of every model and rate, with price, scope, limits and status

GET /api/v1/models needs no key and costs nothing. It lists every model and rate that has not closed, with its price at list, its scope, its limits and its status. The console's Models page and these docs read the same list, so they never disagree. Answers are cached for 60 seconds.

curl -s https://app.makesmores.com/api/v1/models

The answer is {"object": "list", "data": [...]}. One entry looks like this:

{
"id": "ember-v1",
"object": "model",
"slug": "ember-v1",
"name": "Reddit post writer",
"kind": "model",
"group": "writers",
"description": "Writes Reddit posts from a plan of subreddit, angle, evidence and your own specifics, with up to three options to pick from.",
"status": "live",
"async": true,
"mcp_only": false,
"endpoint": "POST /api/v1/drafts",
"endpoints": [
"POST /api/v1/drafts",
"GET /api/v1/drafts/{id}",
"POST /api/v1/drafts/{id}/choice"
],
"mcp_tools": [
"smore_draft_reddit_post",
"smore_get_draft",
"smore_choose_draft"
],
"scope": [
"reddit:posts:write",
"reddit:posts:read"
],
"limits": [
{
"window": "day",
"count": 60,
"per": "workspace",
"unit": "drafts"
},
{
"window": "open",
"count": 5,
"per": "workspace",
"unit": "drafts"
}
],
"billing": "Per token: uncached input, cached input and output, counting every candidate the model writes. The most a draft can cost is deducted when it starts and the rest refunded when it finishes.",
"price": {
"credits_per_1m_input": 5000,
"credits_per_1m_cached_input": 5000,
"credits_per_1m_output": 150000,
"credits_per_item": 0,
"item_unit": null,
"credits_per_request": 0,
"min_credits": 1,
"usd_per_1m_input": 5,
"usd_per_1m_cached_input": 5,
"usd_per_1m_output": 150,
"usd_per_item": null,
"usd_per_request": null,
"usd_min": 0.001,
"credits_per_dollar": 1000,
"effective_from": "2026-10-25T01:57:11Z"
}
}

Fields​

FieldMeaning
id, slugThe public slug, such as ember-v1, or a rate key such as api.call. Both hold the same value
objectAlways model
name, descriptionWhat it does
kindmodel, data or action
groupWhere the console lists it: writers, scorers, security, x_signals or data
statuslive, beta, paused or coming_soon. Closed models leave the list
asynctrue for jobs you submit and then poll: drafts and scans
mcp_onlytrue when only MCP tools reach it, as for the X signals models
endpointThe main REST route, such as POST /api/v1/drafts, or null for MCP-only and coming-soon entries
endpointsEvery REST route that reaches it, including the free status and choice routes
mcp_toolsThe MCP tools that reach it
scopeThe scopes those routes and tools need, as a list
limitsEach {window, count, per, unit}, such as {"window": "day", "count": 60, "per": "workspace", "unit": "drafts"}. window is minute, hour, day, open (at once) or total
billingHow it bills, in a sentence
priceThe rate in force, below, or null

A coming-soon entry has empty endpoints, mcp_tools, scope and limits, price: null, and billing says it is priced before launch.

Price​

FieldMeaning
credits_per_1m_inputCredits per 1M uncached input tokens, or null
credits_per_1m_cached_inputCredits per 1M cached input tokens, or null. When a model has no cached price, this is its input price, since cached input then bills at that price
credits_per_1m_outputCredits per 1M output tokens, or null
credits_per_item, item_unitCredits per item, and what an item is: title, post. item_unit is null when there is no item price
credits_per_requestA fixed fee per successful call
min_creditsThe least a successful call costs
usd_per_1m_input, usd_per_1m_cached_input, usd_per_1m_output, usd_per_item, usd_per_request, usd_minThe same prices in dollars at list, or null where there is no such price
credits_per_dollarCredits per dollar at list
effective_fromWhen this rate took effect

One credit is $0.001, so credits_per_1m_output: 150000 is $150 per 1M output tokens.

Rows that are not models​

idWhat it prices
reddit-mentionsEvery Reddit mentions call, at the per-call price: $0.001 per call
api.callEvery other billable call without its own rate: $0.001 per call
profile.proposalA profile proposal, which reads up to 40 pages: $0.15 per call

All models and prices shows the whole list as a table.