Server definition
- Hash
- sha256:5af221653e2df26c3fef7c5d58376dd8c36b4144dbab008849b986e59d55f73e
- What it is
- What a remote MCP server returned when asked what it offers: 2 tools
The blob, as servednamed by its sha256
{
"instructions": "Answers one question: which is the cheapest LLM that still passes for a task? Call get_verdict with the task in plain words, for example {\"task\":\"spam detection\"}, {\"task\":\"sentiment\"} or {\"task\":\"json extraction\"} (a slug works too): you get the cheapest passing model, its cost per successful task in US dollars with retries and failures counted, the baseline it beats, every candidate, and the page and run file to check it. list_verdicts shows all published verdicts first. Figures are measured on public datasets; the dated price snapshot is in every answer. To get the same verdict on your own examples, open the app_url in the answer.",
"tools": [
{
"description": "Answers: which is the cheapest LLM that still passes for <task>? Give the task in plain words (for example \"spam detection\", \"sentiment\", \"json extraction\", \"tool calling\", \"intent classification\") or a verdict slug, and get the cheapest model that clears the bar, its pass rate and cost per successful task in US dollars, the baseline it is compared with, every candidate's status, plus links to the verdict page, the run file (JSON) and the script that re-runs it. Example: {\"task\":\"spam detection\"} returns the sms-spam verdict. Not sure what exists? Call list_verdicts first.",
"inputSchema": {
"additionalProperties": false,
"properties": {
"slug": {
"description": "Or the exact verdict slug from list_verdicts, for example sms-spam.",
"minLength": 1,
"type": "string"
},
"task": {
"description": "The task in plain words, for example spam detection, sentiment, json extraction. The closest published verdict answers.",
"minLength": 1,
"type": "string"
}
},
"type": "object"
},
"name": "get_verdict",
"outputSchema": {
"properties": {
"answer": {
"description": "One-sentence verdict in plain words, with the price date.",
"type": "string"
},
"app_note": {
"type": "string"
},
"app_url": {
"type": "string"
},
"baseline_cost_per_successful_task_usd": {
"type": [
"number",
"null"
]
},
"baseline_model": {
"type": [
"string",
"null"
]
},
"baseline_pass_rate": {
"type": [
"number",
"null"
]
},
"candidates": {
"items": {
"properties": {
"attempts_per_task": {
"type": [
"number",
"null"
]
},
"cost_per_successful_task_usd": {
"type": [
"number",
"null"
]
},
"latency_p50_ms": {
"type": [
"number",
"null"
]
},
"latency_p95_ms": {
"type": [
"number",
"null"
]
},
"model": {
"type": "string"
},
"pass_rate": {
"type": [
"number",
"null"
]
},
"status": {
"type": [
"string",
"null"
]
}
},
"required": [
"model"
],
"type": "object"
},
"type": "array"
},
"cheapest_passing_model": {
"description": "The cheapest model that clears the bar; null when none does.",
"type": [
"string",
"null"
]
},
"cost_per_successful_task_usd": {
"description": "US dollars per successful task, retries and failures counted.",
"type": [
"number",
"null"
]
},
"dataset": {
"description": "Public dataset behind the run: name, rows, source, licence.",
"type": "object"
},
"matched_by": {
"description": "How the verdict was chosen: the exact slug, or the closest published verdict to the task you described.",
"type": "string"
},
"pass_bar": {
"description": "Pass-rate threshold the cheapest model has to clear (0 to 1).",
"type": "number"
},
"pass_rate": {
"type": [
"number",
"null"
]
},
"price_snapshot_at": {
"type": [
"string",
"null"
]
},
"ran_at": {
"type": [
"string",
"null"
]
},
"rerun_script_url": {
"description": "Script that re-runs the verdict; null when none is published.",
"type": [
"string",
"null"
]
},
"run_file_url": {
"description": "The run file (JSON) to check the figures.",
"type": "string"
},
"saving_pct": {
"type": [
"number",
"null"
]
},
"slug": {
"description": "Verdict slug, the key for get_verdict.",
"type": "string"
},
"task": {
"description": "Task type, for example classification.",
"type": "string"
},
"title": {
"type": "string"
},
"verdict_kind": {
"description": "switch (a cheaper model passes) or keep (the baseline is already cheapest).",
"type": "string"
},
"verdict_url": {
"description": "The verdict page, with the evidence.",
"type": "string"
}
},
"required": [
"slug",
"title",
"task",
"pass_bar",
"cheapest_passing_model",
"cost_per_successful_task_usd",
"baseline_model",
"verdict_url",
"answer",
"verdict_kind",
"candidates",
"run_file_url",
"app_url"
],
"type": "object"
}
},
{
"description": "Which LLM is the cheapest that still passes? Lists every published Clearbar verdict at once: the task, the pass bar, the cheapest passing model, its cost per successful task (retries and failures counted) and the page to check it. Optional filter by task type. Next call: get_verdict with {\"task\":\"spam detection\"} (or a slug from this list) for the full table.",
"inputSchema": {
"additionalProperties": false,
"properties": {
"task": {
"description": "Only verdicts of this task type, for example classification, extraction or tool-call.",
"type": "string"
}
},
"type": "object"
},
"name": "list_verdicts",
"outputSchema": {
"properties": {
"app_note": {
"type": "string"
},
"app_url": {
"description": "Where to run the same verdict on your own examples.",
"type": "string"
},
"count": {
"description": "Number of verdicts returned.",
"type": "integer"
},
"next_call": {
"description": "The call to make next for the full table of one verdict.",
"type": "string"
},
"prices_as_of": {
"description": "Date (YYYY-MM-DD) of the newest price snapshot behind the figures.",
"type": [
"string",
"null"
]
},
"verdicts": {
"items": {
"properties": {
"baseline_model": {
"type": [
"string",
"null"
]
},
"cheapest_passing_model": {
"description": "The cheapest model that clears the bar; null when none does.",
"type": [
"string",
"null"
]
},
"cost_per_successful_task_usd": {
"description": "US dollars per successful task, retries and failures counted.",
"type": [
"number",
"null"
]
},
"pass_bar": {
"description": "Pass-rate threshold the cheapest model has to clear (0 to 1).",
"type": "number"
},
"pass_rate": {
"type": [
"number",
"null"
]
},
"slug": {
"description": "Verdict slug, the key for get_verdict.",
"type": "string"
},
"task": {
"description": "Task type, for example classification.",
"type": "string"
},
"title": {
"type": "string"
},
"verdict_url": {
"description": "The verdict page, with the evidence.",
"type": "string"
}
},
"required": [
"slug",
"title",
"task",
"pass_bar",
"cheapest_passing_model",
"cost_per_successful_task_usd",
"baseline_model",
"verdict_url"
],
"type": "object"
},
"type": "array"
}
},
"required": [
"count",
"prices_as_of",
"verdicts",
"app_url",
"app_note"
],
"type": "object"
}
}
]
}Verify it yourself
curl -s https://api.teppi.xyz/v1/evidence/sha256:5af221653e2df26c3fef7c5d58376dd8c36b4144dbab008849b986e59d55f73e | sha256sum