Tool tree
Evaluate and compare how well large language models solve Nonogram (Picross) puzzles. Interactive benchmark results, visualizations, and leaderboards for AI reasoning capabilities.
Implementation
gradeNot graded
gradeNot graded
page: Every page
get_leaderboardAnswer
Nonobench models ranked by accuracy, overall or for one grid size. Effort defaults to all levels; use effort=best to match the homepage.
View tool JSON
{
"name": "get_leaderboard",
"kind": "answer",
"impl": "imperative",
"description": "Nonobench models ranked by accuracy, overall or for one grid size. Effort defaults to all levels; use effort=best to match the homepage.",
"inputSchema": {
"type": "object",
"properties": {
"size": {
"type": "string",
"enum": [
"5x5",
"10x10",
"15x15",
"20x20"
],
"description": "Grid size to filter on"
},
"provider": {
"type": "string",
"description": "Comma-separated provider ids"
},
"family": {
"type": "string",
"description": "Comma-separated family ids"
},
"version": {
"type": "string",
"description": "Comma-separated benchmark versions: 1.0, 1.1, 1.2"
},
"effort": {
"type": "string",
"description": "best, all, or an effort level"
},
"reasoning": {
"type": "boolean"
},
"open_weights": {
"type": "boolean"
}
}
},
"page": "/"
}list_providersAnswer
Provider ids, names, families and variant counts.
View tool JSON
{
"name": "list_providers",
"kind": "answer",
"impl": "imperative",
"description": "Provider ids, names, families and variant counts.",
"inputSchema": {
"type": "object",
"properties": {}
},
"page": "/"
}list_familiesAnswer
Model families, efforts and best variants.
View tool JSON
{
"name": "list_families",
"kind": "answer",
"impl": "imperative",
"description": "Model families, efforts and best variants.",
"inputSchema": {
"type": "object",
"properties": {}
},
"page": "/"
}compare_modelsAction
Compare two or more model ids or family names side by side.
View tool JSON
{
"name": "compare_models",
"kind": "act",
"impl": "imperative",
"description": "Compare two or more model ids or family names side by side.",
"inputSchema": {
"type": "object",
"properties": {
"models": {
"type": "array",
"items": {
"type": "string"
},
"minItems": 2
}
},
"required": [
"models"
]
},
"page": "/"
}set_filtersAction
Update the visible leaderboard filters and URL.
View tool JSON
{
"name": "set_filters",
"kind": "act",
"impl": "imperative",
"description": "Update the visible leaderboard filters and URL.",
"inputSchema": {
"type": "object",
"properties": {
"size": {
"type": "string",
"enum": [
"5x5",
"10x10",
"15x15",
"20x20"
],
"description": "Grid size to filter on"
},
"provider": {
"type": "string",
"description": "Comma-separated provider ids"
},
"family": {
"type": "string",
"description": "Comma-separated family ids"
},
"version": {
"type": "string",
"description": "Comma-separated benchmark versions: 1.0, 1.1, 1.2"
},
"effort": {
"type": "string",
"description": "best, all, or an effort level"
},
"reasoning": {
"type": "boolean"
},
"open_weights": {
"type": "boolean"
}
}
},
"page": "/"
}get_model_resultsAnswer
Accuracy, cost and latency for one model, per grid size.
View tool JSON
{
"name": "get_model_results",
"kind": "answer",
"impl": "imperative",
"description": "Accuracy, cost and latency for one model, per grid size.",
"inputSchema": {
"type": "object",
"properties": {
"model": {
"type": "string",
"description": "Model name from the leaderboard"
}
},
"required": [
"model"
]
},
"page": "/"
}list_puzzlesAnswer
The benchmark puzzles with ids and row/column clues.
View tool JSON
{
"name": "list_puzzles",
"kind": "answer",
"impl": "imperative",
"description": "The benchmark puzzles with ids and row/column clues.",
"inputSchema": {
"type": "object",
"properties": {
"size": {
"type": "string",
"enum": [
"5x5",
"10x10",
"15x15",
"20x20"
],
"description": "Grid size to filter on"
}
}
},
"page": "/"
}get_puzzle_resultsAnswer
Per-model outcomes for a puzzle, optionally including parsed grids.
View tool JSON
{
"name": "get_puzzle_results",
"kind": "answer",
"impl": "imperative",
"description": "Per-model outcomes for a puzzle, optionally including parsed grids.",
"inputSchema": {
"type": "object",
"properties": {
"id": {
"type": "string"
},
"size": {
"type": "string",
"enum": [
"5x5",
"10x10",
"15x15",
"20x20"
],
"description": "Grid size to filter on"
},
"provider": {
"type": "string",
"description": "Comma-separated provider ids"
},
"family": {
"type": "string",
"description": "Comma-separated family ids"
},
"version": {
"type": "string",
"description": "Comma-separated benchmark versions: 1.0, 1.1, 1.2"
},
"effort": {
"type": "string",
"description": "best, all, or an effort level"
},
"reasoning": {
"type": "boolean"
},
"open_weights": {
"type": "boolean"
},
"include_answers": {
"type": "boolean"
}
},
"required": [
"id"
]
},
"page": "/"
}get_model_puzzlesAnswer
Which puzzles a model solved, missed, or did not run.
View tool JSON
{
"name": "get_model_puzzles",
"kind": "answer",
"impl": "imperative",
"description": "Which puzzles a model solved, missed, or did not run.",
"inputSchema": {
"type": "object",
"properties": {
"model": {
"type": "string"
}
},
"required": [
"model"
]
},
"page": "/"
}check_solutionAnswer
Check a nonogram grid (row-major 0/1 string) against a puzzle's clues.
View tool JSON
{
"name": "check_solution",
"kind": "answer",
"impl": "imperative",
"description": "Check a nonogram grid (row-major 0/1 string) against a puzzle's clues.",
"inputSchema": {
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Puzzle id"
},
"grid": {
"type": "string"
}
},
"required": [
"id",
"grid"
]
},
"page": "/"
}open_puzzleAction
Show a puzzle in the puzzle explorer on this page.
View tool JSON
{
"name": "open_puzzle",
"kind": "act",
"impl": "imperative",
"description": "Show a puzzle in the puzzle explorer on this page.",
"inputSchema": {
"type": "object",
"properties": {
"index": {
"type": "integer",
"minimum": 0,
"maximum": 39,
"description": "0-based puzzle index from list_puzzles"
},
"model": {
"type": "string",
"description": "Optional model variant to show its answer overlay"
}
},
"required": [
"index"
]
},
"page": "/"
}