{
  "model_name": "GLM-5.2",
  "model_organization": "Z.ai",
  "submitting_organization": "Sierra",
  "submission_date": "2026-08-04",
  "submission_type": "standard",
  "modality": "text",
  "contact_info": {
    "email": "ben.s@sierra.ai",
    "name": "Sierra Research Team"
  },
  "results": {
    "airline": null,
    "retail": null,
    "telecom": null,
    "banking_knowledge": {
      "pass_1": 37.11340206185567,
      "pass_2": 25.257731958762886,
      "pass_3": 18.298969072164947,
      "pass_4": 13.402061855670103,
      "cost": null,
      "retrieval_config": "alltools"
    }
  },
  "is_new": true,
  "trajectories_available": true,
  "trajectory_files": {
    "banking_knowledge": "banking_knowledge_results.json"
  },
  "references": [],
  "reasoning_effort": "xhigh",
  "methodology": {
    "evaluation_date": "2026-07-24",
    "tau2_bench_version": "1.0.1",
    "user_simulator": "gpt-5.2",
    "notes": "Banking_knowledge evaluation with AllTools retrieval, GPT-5.2 low-reasoning user simulation, four trials, and seed 300.",
    "verification": {
      "modified_prompts": false,
      "omitted_questions": false
    }
  },
  "model_release": {
    "release_date": "2026-06-16",
    "announcement_url": "https://z.ai/blog/glm-5.2",
    "announcement_title": "GLM-5.2: Built for Long-Horizon Tasks"
  }
}