Cheapest LLM that passes each task: cost per successful task and pass rate on public datasets.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
The full verdict for one task: the cheapest model that clears the bar, its pass rate and cost per successful task in US dollars, the baseline model 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. Get the slug from list_verdicts.
Verdict slug from list_verdicts, for example sms-spam.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http verdicts https://clearbar.grain64.com/mcp
The full verdict for one task: the cheapest model that clears the bar, its pass rate and cost per successful task in US dollars, the baseline model 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. Get the slug from list_verdicts.
| Argument | Type | Description |
|---|---|---|
| slug* | string | Verdict slug from list_verdicts, for example sms-spam. |
{
"type": "object",
"required": [
"slug"
],
"properties": {
"slug": {
"type": "string",
"description": "Verdict slug from list_verdicts, for example sms-spam."
}
},
"additionalProperties": false
}Which LLM is the cheapest that still passes? Lists every published Clearbar verdict: 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.
| Argument | Type | Description |
|---|---|---|
| task | string | Only verdicts of this task type, for example classification, extraction or tool-call. |
{
"type": "object",
"properties": {
"task": {
"type": "string",
"description": "Only verdicts of this task type, for example classification, extraction or tool-call."
}
},
"additionalProperties": false
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 6, 2026.