{
  "name": "TensionLM-117M-TS-Reasoner-v10",
  "repo_id": "BoggersTheFish/TensionLM-117M-TS-Reasoner-v10",
  "artifact_type": "broadened_interactive_cpu_ts_trace_reasoner",
  "fallback_model": "BoggersTheFish/TensionLM-117M-Reasoning-v2",
  "supported_families": [
    "transitivity",
    "arithmetic",
    "code_reasoning",
    "boolean_logic",
    "set_reasoning",
    "string_reasoning"
  ],
  "fixed_benchmarks": {
    "tac_v2": {
      "prefix_correct": 120,
      "substring_correct": 120,
      "total": 120,
      "solve_rate": 1.0
    },
    "tac_v3": {
      "prefix_correct": 120,
      "substring_correct": 120,
      "total": 120,
      "solve_rate": 1.0
    },
    "tac_v4": {
      "prefix_correct": 120,
      "substring_correct": 120,
      "total": 120,
      "solve_rate": 1.0
    }
  },
  "receipts": {
    "examples_trace": {
      "correct": 30,
      "total": 30,
      "accuracy": 1.0
    },
    "server_smoke": {
      "ok": true
    },
    "new_standard": {
      "prefix_correct": 3000,
      "total": 3000,
      "solve_rate": 1.0
    },
    "new_paraphrase": {
      "prefix_correct": 3000,
      "total": 3000,
      "solve_rate": 1.0
    },
    "new_unknown": {
      "prefix_correct": 3000,
      "total": 3000,
      "solve_rate": 0.0
    },
    "new_mixed": {
      "prefix_correct": 3000,
      "total": 3000,
      "solve_rate": 0.6723333333333333
    },
    "all_mixed": {
      "prefix_correct": 6000,
      "total": 6000,
      "solve_rate": 0.6608333333333334
    }
  },
  "claim_boundary": "Broadened interactive CPU trace API over generated formal prompt families; unknown correctness means explainable abstention."
}