{
  "updatedAt": "2026-06-04T03:17:32.400Z",
  "source": "openrouter-models-api",
  "rankingVersion": "2026-06-04.v1",
  "probeMode": "active",
  "refreshMode": "fullEval",
  "liteEvalMode": "active",
  "liteEvalSuite": "lite-agent-eval-v1",
  "rankingConfidence": "high",
  "rankingConfidenceReason": "No unevaluated candidate can overtake the primary even with a perfect lite eval score.",
  "evalCoverage": {
    "evaluated": 4,
    "completedToday": 2,
    "attemptedToday": 4,
    "totalRanked": 20,
    "candidatePool": 20,
    "unevaluatedCanStillWin": false,
    "unevaluatedThreatCount": 0,
    "topUnevaluatedThreat": null,
    "maxUnevaluatedPossibleScore": null
  },
  "baseUrl": "https://openrouter.ai/api/v1",
  "createKeyUrl": "https://openrouter.ai/settings/keys",
  "fallback": {
    "id": "openrouter/free",
    "reason": "OpenRouter-managed free model router"
  },
  "count": 6,
  "models": [
    {
      "rank": 1,
      "id": "nvidia/nemotron-3-super-120b-a12b:free",
      "name": "NVIDIA: Nemotron 3 Super (free)",
      "score": 1390,
      "metadataScore": 585,
      "healthScore": 250,
      "latencyScore": 60,
      "liteEvalScore": 495,
      "contextLength": 262144,
      "maxCompletionTokens": 262144,
      "supportsTools": true,
      "supportsToolChoice": true,
      "supportsStructuredOutputs": true,
      "supportsResponseFormat": true,
      "supportsReasoning": true,
      "supportsIncludeReasoning": true,
      "supportsSeed": true,
      "supportsStop": false,
      "latencyMs": 1753,
      "healthStatus": "imperfect",
      "evalSuite": "lite-agent-eval-v1",
      "evalSummary": {
        "suite": "lite-agent-eval-v1",
        "status": "completed",
        "passed": 1,
        "total": 3,
        "updatedAt": "2026-06-04T03:17:32.400Z",
        "tasks": [
          {
            "id": "task_files_lite",
            "source": "pinchbench/task_files",
            "points": 220,
            "maxPoints": 350,
            "score": 0.629,
            "status": "partial",
            "details": {
              "usedTool": true,
              "mainPyCreated": true,
              "helloPython": true,
              "readmeCreated": false,
              "readmeHasTitle": false,
              "gitignoreCreated": false,
              "gitignoreHasPycache": false,
              "noUnsafePaths": true
            }
          },
          {
            "id": "task_shell_command_lite",
            "source": "pinchbench/task_shell_command_generator",
            "points": 275,
            "maxPoints": 300,
            "score": 0.917,
            "status": "passed",
            "details": {
              "command": "grep -rl \"FATAL:\" . --include=\"*.log\"",
              "wroteCommandFile": true,
              "shellOnly": true,
              "recursiveSearch": true,
              "restrictsToLogFiles": true,
              "matchesFatalExactly": true,
              "printsEachFileOnce": false,
              "plausiblyExecutable": true
            }
          },
          {
            "id": "babylonian_clock_lite",
            "source": "shir-man/custom",
            "points": 0,
            "maxPoints": 100,
            "score": 0,
            "status": "failed",
            "details": {
              "expected": "02:12:48",
              "received": "{\". We need to parse the Babylonian-style clock: symbols: 𒁹 = 1, 𒌋 = 10. There"
            }
          }
        ]
      },
      "reason": "Tools, structured outputs, reasoning; lite eval 1/3",
      "instabilityPenalty": 0
    },
    {
      "rank": 2,
      "id": "google/gemma-4-31b-it:free",
      "name": "Google: Gemma 4 31B (free)",
      "score": 1210,
      "metadataScore": 565,
      "healthScore": 100,
      "latencyScore": 20,
      "liteEvalScore": 625,
      "contextLength": 262144,
      "maxCompletionTokens": 32768,
      "supportsTools": true,
      "supportsToolChoice": true,
      "supportsStructuredOutputs": false,
      "supportsResponseFormat": true,
      "supportsReasoning": true,
      "supportsIncludeReasoning": true,
      "supportsSeed": true,
      "supportsStop": true,
      "latencyMs": 5525,
      "healthStatus": "http_429",
      "evalSuite": "lite-agent-eval-v1",
      "evalSummary": {
        "suite": "lite-agent-eval-v1",
        "status": "model_rate_limited",
        "passed": 2,
        "total": 3,
        "updatedAt": "2026-06-04T03:17:32.400Z",
        "tasks": [
          {
            "id": "task_files_lite",
            "source": "pinchbench/task_files",
            "points": 350,
            "maxPoints": 350,
            "score": 1,
            "status": "passed",
            "details": {
              "usedTool": true,
              "mainPyCreated": true,
              "helloPython": true,
              "readmeCreated": true,
              "readmeHasTitle": true,
              "gitignoreCreated": true,
              "gitignoreHasPycache": true,
              "noUnsafePaths": true
            }
          },
          {
            "id": "task_shell_command_lite",
            "source": "pinchbench/task_shell_command_generator",
            "points": 275,
            "maxPoints": 300,
            "score": 0.917,
            "status": "passed",
            "details": {
              "command": "grep -rl \"FATAL:\" --include=\"*.log\" .",
              "wroteCommandFile": true,
              "shellOnly": true,
              "recursiveSearch": true,
              "restrictsToLogFiles": true,
              "matchesFatalExactly": true,
              "printsEachFileOnce": false,
              "plausiblyExecutable": true
            }
          },
          {
            "id": "babylonian_clock_lite",
            "source": "shir-man/custom",
            "points": 0,
            "maxPoints": 100,
            "score": 0,
            "status": "model_rate_limited",
            "details": {
              "responseStatus": 429,
              "rateLimitKind": "model_or_provider_limit",
              "providerName": "Google AI Studio",
              "rateLimitLimit": null,
              "rateLimitRemaining": null,
              "rateLimitResetAt": null,
              "responsePreview": "{\"error\":{\"message\":\"Provider returned error\",\"code\":429,\"metadata\":{\"raw\":\"google/gemma-4-31b-it:free is temporarily rate-limited upstream. Please retry shortl"
            }
          }
        ]
      },
      "reason": "Tools, reasoning, 262k context",
      "instabilityPenalty": 100
    },
    {
      "rank": 3,
      "id": "openrouter/owl-alpha",
      "name": "Owl Alpha",
      "score": 655,
      "metadataScore": 530,
      "healthScore": 100,
      "latencyScore": 25,
      "liteEvalScore": 0,
      "contextLength": 1048756,
      "maxCompletionTokens": 262144,
      "supportsTools": true,
      "supportsToolChoice": false,
      "supportsStructuredOutputs": true,
      "supportsResponseFormat": true,
      "supportsReasoning": false,
      "supportsIncludeReasoning": false,
      "supportsSeed": true,
      "supportsStop": true,
      "latencyMs": null,
      "healthStatus": "not_probed",
      "evalSuite": "lite-agent-eval-v1",
      "evalSummary": {
        "suite": "lite-agent-eval-v1",
        "status": "completed",
        "passed": 0,
        "total": 3,
        "updatedAt": "2026-06-04T03:17:32.400Z",
        "tasks": [
          {
            "id": "task_files_lite",
            "source": "pinchbench/task_files",
            "points": 0,
            "maxPoints": 350,
            "score": 0,
            "status": "http_timeout_or_error",
            "details": {
              "responseStatus": "timeout_or_error",
              "rateLimitKind": "none",
              "providerName": null,
              "rateLimitLimit": null,
              "rateLimitRemaining": null,
              "rateLimitResetAt": null,
              "responsePreview": "The operation was aborted"
            }
          },
          {
            "id": "task_shell_command_lite",
            "source": "pinchbench/task_shell_command_generator",
            "points": 0,
            "maxPoints": 300,
            "score": 0,
            "status": "http_timeout_or_error",
            "details": {
              "responseStatus": "timeout_or_error",
              "rateLimitKind": "none",
              "providerName": null,
              "rateLimitLimit": null,
              "rateLimitRemaining": null,
              "rateLimitResetAt": null,
              "responsePreview": "The operation was aborted"
            }
          },
          {
            "id": "babylonian_clock_lite",
            "source": "shir-man/custom",
            "points": 0,
            "maxPoints": 100,
            "score": 0,
            "status": "http_timeout_or_error",
            "details": {
              "responseStatus": "timeout_or_error",
              "rateLimitKind": "none",
              "providerName": null,
              "rateLimitLimit": null,
              "rateLimitRemaining": null,
              "rateLimitResetAt": null,
              "responsePreview": "The operation was aborted"
            }
          }
        ]
      },
      "reason": "Tools, structured outputs, 262k context; lite eval 0/3",
      "instabilityPenalty": 0
    },
    {
      "rank": 4,
      "id": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free",
      "name": "NVIDIA: Nemotron 3 Nano Omni (free)",
      "score": 619,
      "metadataScore": 494,
      "healthScore": 100,
      "latencyScore": 25,
      "liteEvalScore": 0,
      "contextLength": 256000,
      "maxCompletionTokens": 65536,
      "supportsTools": true,
      "supportsToolChoice": true,
      "supportsStructuredOutputs": false,
      "supportsResponseFormat": false,
      "supportsReasoning": true,
      "supportsIncludeReasoning": true,
      "supportsSeed": true,
      "supportsStop": false,
      "latencyMs": null,
      "healthStatus": "not_probed",
      "evalSuite": "lite-agent-eval-v1",
      "evalSummary": {
        "suite": "lite-agent-eval-v1",
        "status": "not_evaluated",
        "passed": 0,
        "total": 0,
        "updatedAt": null,
        "tasks": []
      },
      "reason": "Tools, reasoning, 128k+ context",
      "instabilityPenalty": 0
    },
    {
      "rank": 5,
      "id": "poolside/laguna-m.1:free",
      "name": "Poolside: Laguna M.1 (free)",
      "score": 600,
      "metadataScore": 475,
      "healthScore": 100,
      "latencyScore": 25,
      "liteEvalScore": 0,
      "contextLength": 262144,
      "maxCompletionTokens": 32768,
      "supportsTools": true,
      "supportsToolChoice": true,
      "supportsStructuredOutputs": false,
      "supportsResponseFormat": false,
      "supportsReasoning": true,
      "supportsIncludeReasoning": true,
      "supportsSeed": false,
      "supportsStop": false,
      "latencyMs": null,
      "healthStatus": "not_probed",
      "evalSuite": "lite-agent-eval-v1",
      "evalSummary": {
        "suite": "lite-agent-eval-v1",
        "status": "not_evaluated",
        "passed": 0,
        "total": 0,
        "updatedAt": null,
        "tasks": []
      },
      "reason": "Tools, reasoning, 262k context",
      "instabilityPenalty": 0
    },
    {
      "rank": 6,
      "id": "poolside/laguna-xs.2:free",
      "name": "Poolside: Laguna XS.2 (free)",
      "score": 600,
      "metadataScore": 475,
      "healthScore": 100,
      "latencyScore": 25,
      "liteEvalScore": 0,
      "contextLength": 262144,
      "maxCompletionTokens": 32768,
      "supportsTools": true,
      "supportsToolChoice": true,
      "supportsStructuredOutputs": false,
      "supportsResponseFormat": false,
      "supportsReasoning": true,
      "supportsIncludeReasoning": true,
      "supportsSeed": false,
      "supportsStop": false,
      "latencyMs": null,
      "healthStatus": "not_probed",
      "evalSuite": "lite-agent-eval-v1",
      "evalSummary": {
        "suite": "lite-agent-eval-v1",
        "status": "not_evaluated",
        "passed": 0,
        "total": 0,
        "updatedAt": null,
        "tasks": []
      },
      "reason": "Tools, reasoning, 262k context",
      "instabilityPenalty": 0
    }
  ],
  "notes": [
    "Free model availability and rate limits can vary.",
    "Top candidates also received a lite agent eval for small file-writing, shell-command, and symbolic decoding tasks.",
    "The lite eval is a practical heuristic, not an official PinchBench result.",
    "Use openrouter/free if the recommended model is temporarily unavailable."
  ]
}