Cheap First is an MCP server published by io.github.robd-hub, hosted at cheap-first.vercel.app. When last checked on 2026-11-21 it was working: it answers and lists its tools. It declares 2 tools, including ask_cheap_first, cheap_first_stats.
Answers cheaply first, then escalates to a premium model only when a checker fails.
87 connections through 5 shared values. Tap any node.
Squares are this server's values; dots are other servers sharing them. A shared address or publisher is a lead, not proof of a common operator.
Inferred from the host name and headers: what a server runs on, not who runs it.
Answer a question the Cheap First way: Dolphin answers, another cheap sample checks numbers and choices, SERV Reasoning grades anything else, and a premium model answers instead only if the check misses the pass mark. Returns the answer, whether it escalated, the check score, the cost and the estimated saving. Free calls use the public demo limits (600 characters, 10 an hour per IP, 300 a day). Send Authorization: Bearer cf_... or an x402 payment of $0.005 USDC on Base for the paid limits (2000 characters). To receive the x402 challenge, send the header X-Cheap-First-Paid: 1. mode: strict escalates more often (pass mark 85), balanced is the default (75), budget escalates less often (60).
| Parameter | Type | Description |
|---|---|---|
| mode | string | strict, balanced (default) or budget. Unknown values are treated as balanced. |
| question * | string | The question. 600 characters on the free demo, 2000 when paid. |
{
"name": "ask_cheap_first",
"title": "Ask Cheap First",
"description": "Answer a question the Cheap First way: Dolphin answers, another cheap sample checks numbers and choices, SERV Reasoning grades anything else, and a premium model answers instead only if the check misses the pass mark. Returns the answer, whether it escalated, the check score, the cost and the estimated saving. Free calls use the public demo limits (600 characters, 10 an hour per IP, 300 a day). Send Authorization: Bearer cf_... or an x402 payment of $0.005 USDC on Base for the paid limits (2000 characters). To receive the x402 challenge, send the header X-Cheap-First-Paid: 1. mode: strict escalates more often (pass mark 85), balanced is the default (75), budget escalates less often (60).",
"inputSchema": {
"type": "object",
"required": [
"question"
],
"properties": {
"mode": {
"enum": [
"strict",
"balanced",
"budget"
],
"type": "string",
"description": "strict, balanced (default) or budget. Unknown values are treated as balanced."
},
"question": {
"type": "string",
"description": "The question. 600 characters on the free demo, 2000 when paid."
}
}
},
"outputSchema": {
"type": "object",
"required": [
"answer",
"escalated",
"checkScore",
"costUsd",
"savedUsd"
],
"properties": {
"mode": {
"enum": [
"strict",
"balanced",
"budget"
],
"type": "string"
},
"tier": {
"enum": [
"free",
"paid"
],
"type": "string"
},
"model": {
"type": "string",
"description": "Label of the model that produced the answer."
},
"answer": {
"type": "string",
"description": "The answer to give the user."
},
"costUsd": {
"type": "number",
"description": "What Cheap First spent on this call, in US dollars."
},
"paidUsd": {
"type": "number",
"description": "What the caller paid, on the paid tier."
},
"savedUsd": {
"type": "number",
"description": "Estimated saving against sending the question to the premium model alone. Can be negative when an escalation cost more than that."
},
"escalated": {
"type": "boolean",
"description": "True when the check failed and the premium model answered instead."
},
"answeredBy": {
"enum": [
"cheap",
"premium"
],
"type": "string",
"description": "cheap is Dolphin. premium is the frontier model."
},
"checkScore": {
"type": "number",
"description": "The checker's score for the cheap answer, from 0 to 100."
},
"checkReason": {
"type": "string",
"description": "Why the check passed or failed, in one line."
},
"checkSignal": {
"enum": [
"agreement",
"arithmetic",
"grader",
"blank"
],
"type": "string",
"description": "agreement: extra cheap samples. arithmetic: an equation did not evaluate. grader: the language-model grade. blank: Dolphin returned nothing."
},
"creditsLeft": {
"type": "number",
"description": "Prepaid answers left on the key, when one was used."
},
"transaction": {
"type": "string",
"description": "x402 settlement transaction, when there is one."
}
},
"description": "One Cheap First answer, plus what the check decided and what it cost."
},
"annotations": {
"readOnlyHint": false,
"openWorldHint": true,
"destructiveHint": false
}
}Published benchmark figures from the saved result files. The in-house run, the harder run graded by code, and the public GSM8K and ARC-Challenge run when that file is complete. The public summary includes accuracy, cost, escalation, latency, and where the checker missed. Does not call a model.
No parameters.
{
"name": "cheap_first_stats",
"title": "Cheap First benchmark",
"description": "Published benchmark figures from the saved result files. The in-house run, the harder run graded by code, and the public GSM8K and ARC-Challenge run when that file is complete. The public summary includes accuracy, cost, escalation, latency, and where the checker missed. Does not call a model.",
"inputSchema": {
"type": "object",
"properties": {},
"additionalProperties": false
},
"outputSchema": {
"type": "object",
"required": [
"benchmark",
"hardBenchmark",
"publicBenchmark",
"resultsPage"
],
"properties": {
"benchmark": {
"type": "object",
"description": "The in-house benchmark, with costs rounded to cents the way the site quotes them, plus exactUsd."
},
"resultsPage": {
"type": "string"
},
"hardBenchmark": {
"type": [
"object",
"null"
],
"description": "The harder benchmark graded by code, or null if it was not built."
},
"publicBenchmark": {
"type": [
"object",
"null"
],
"description": "The public GSM8K and ARC-Challenge run graded by code, or null if that file is missing or incomplete. summary includes the cost, the accuracy gap, and the checker's misses."
}
}
},
"annotations": {
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Cheap First answers with Dolphin's cheap model. A number or a choice is checked by another cheap sample and by evaluating any equation; anything else is scored by SERV Reasoning. A premium model is called only when that check misses the pass mark. Use ask_cheap_first for a question and cheap_first_stats for the published benchmark. Free calls share the public demo limits: 600 characters, 10 an hour from one IP, 300 a day, and the demo's daily premium budget. A prepaid key (Authorization: Bearer cf_...) uses the paid API, one credit per answer. An x402 payment of $0.005 USDC on Base does too: send X-Cheap-First-Paid: 1 and the endpoint replies 402 with the usual challenge. mode is strict (pass mark 85), balanced (75, the default) or budget (60).
Names, descriptions and schemas were written by the server and are shown as data.
Pattern matches on what the server declares. They show what it says it can do, not what it does.
None yet. Changes appear after a server's second daily check.