{
  "level1": [
    {
      "agent": "MopMonk Agent",
      "model": "MiniMax M3",
      "model_release_date": "2026-06-01",
      "include_in_plot": false,
      "score_10": 0.731,
      "date": "2026-06-28",
      "source": "MopMonk AI",
      "source_url": "https://github.com/MopMonkAI/MopMonkAgent",
      "trials": 1,
      "focus": "agent"
    },
    {
      "agent": "OpenAI Agent",
      "model": "GPT-5.5-Cyber",
      "model_release_date": "2026-06-22",
      "include_in_plot": true,
      "score_10": 0.856,
      "date": "2026-06-22",
      "source": "OpenAI",
      "source_url": "https://openai.com/index/daybreak-securing-the-world/",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Crystalline (with a pre-seeded, test-time-updated knowledge base)",
      "model": "Claude Opus 4.6",
      "model_release_date": "2026-02-05",
      "include_in_plot": true,
      "icon": "/assets/images/cybergym/crystalline.png",
      "features": ["Knowledge base", "Test-time memory"],
      "score_10": 0.896,
      "date": "2026-06-08",
      "source": "Independent researcher",
      "source_url": "https://github.com/synchopate/cybergym-logos",
      "trials": 1,
      "focus": "agent"
    },
    {
      "agent": "MDASH",
      "model": "Multi-model (GPT-5.4, Claude Opus 4.6, Claude Sonnet 4.6)",
      "model_release_date": "2026-05-12",
      "include_in_plot": true,
      "plot_links": ["GPT-5.4", "Claude Opus 4.6", "Claude Sonnet 4.6"],
      "features": ["Multi-model", "Orchestration"],
      "score_10": 0.8845,
      "date": "2026-05-12",
      "source": "Microsoft",
      "source_url": "https://www.microsoft.com/en-us/security/blog/2026/05/12/defense-at-ai-speed-microsofts-new-multi-model-agentic-security-system-tops-leading-industry-benchmark/",
      "trials": 1,
      "focus": "agent"
    },
    {
      "agent": "Meta Agent",
      "model": "Muse Spark",
      "model_release_date": "2026-04-08",
      "include_in_plot": true,
      "score_10": 0.435,
      "date": "2026-04-14",
      "source": "Meta AI",
      "source_url": "https://ai.meta.com/static-resource/muse-spark-safety-and-preparedness-report",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenAI Agent",
      "model": "GPT-5.4",
      "model_release_date": "2026-03-05",
      "include_in_plot": true,
      "score_10": 0.79,
      "date": "2026-04-23",
      "source": "OpenAI",
      "source_url": "https://openai.com/index/introducing-gpt-5-5",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenAI Agent",
      "model": "GPT-5.5",
      "model_release_date": "2026-04-23",
      "include_in_plot": true,
      "score_10": 0.818,
      "date": "2026-04-23",
      "source": "OpenAI",
      "source_url": "https://openai.com/index/introducing-gpt-5-5",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Claude Code",
      "model": "GLM-5",
      "model_release_date": "2026-02-12",
      "include_in_plot": true,
      "score_10": 0.432,
      "date": "2026-02-12",
      "source": "Zhipu AI",
      "source_url": "https://z.ai/blog/glm-5",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Claude Code",
      "model": "GLM-5.1",
      "model_release_date": "2026-04-07",
      "include_in_plot": true,
      "score_10": 0.687,
      "date": "2026-04-07",
      "source": "Zhipu AI",
      "source_url": "https://z.ai/blog/glm-5.1",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Codex CLI",
      "model": "GPT-5.4",
      "model_release_date": "2026-03-05",
      "include_in_plot": false,
      "score_10": 0.663,
      "date": "2026-04-07",
      "source": "Zhipu AI",
      "source_url": "https://z.ai/blog/glm-5.1",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Gemini CLI",
      "model": "Gemini 3.1 Pro",
      "model_release_date": "2026-02-19",
      "include_in_plot": false,
      "score_10": 0.388,
      "date": "2026-04-07",
      "source": "Zhipu AI",
      "source_url": "https://z.ai/blog/glm-5.1",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Claude Code",
      "model": "GLM-4.7",
      "model_release_date": "2025-12-22",
      "include_in_plot": true,
      "score_10": 0.235,
      "date": "2026-02-12",
      "source": "Zhipu AI",
      "source_url": "https://z.ai/blog/glm-5",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Kimi Agent",
      "model": "Kimi K2.5",
      "model_release_date": "2026-02-02",
      "include_in_plot": true,
      "score_10": 0.413,
      "date": "2026-02-02",
      "source": "Kimi",
      "source_url": "https://arxiv.org/abs/2602.02276",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "SageAgent",
      "model": "GPT-5",
      "model_release_date": "2025-08-07",
      "include_in_plot": true,
      "icon": "/assets/images/cybergym/sageagent.png",
      "plot_links": ["GPT-5 (high)"],
      "score_10": 0.602,
      "date": "2026-02-09",
      "source": "OpenSage Team",
      "source_url": "https://www.opensage-agent.ai/",
      "trials": 1,
      "focus": "agent"
    },
    {
      "agent": "Anthropic Agent",
      "model": "Claude Mythos Preview",
      "model_release_date": "2026-04-07",
      "include_in_plot": true,
      "score_10": 0.831,
      "date": "2026-04-07",
      "source": "Anthropic",
      "source_url": "https://www.anthropic.com/claude-mythos-preview-system-card",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Anthropic Agent",
      "model": "Claude Opus 4.6",
      "model_release_date": "2026-02-05",
      "include_in_plot": true,
      "score_10": 0.666,
      "date": "2026-02-05",
      "source": "Anthropic",
      "source_url": "https://www.anthropic.com/claude-opus-4-6-system-card",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Anthropic Agent",
      "model": "Claude Sonnet 4.6",
      "model_release_date": "2026-02-17",
      "include_in_plot": true,
      "score_10": 0.652,
      "date": "2026-02-17",
      "source": "Anthropic",
      "source_url": "https://www.anthropic.com/claude-sonnet-4-6-system-card",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Anthropic Agent",
      "model": "Claude Opus 4.5",
      "model_release_date": "2025-11-24",
      "include_in_plot": true,
      "score_10": 0.5063,
      "date": "2026-02-05",
      "source": "Anthropic",
      "source_url": "https://www.anthropic.com/claude-opus-4-5-system-card",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Anthropic Agent",
      "model": "Claude Sonnet 4.5",
      "model_release_date": "2025-09-29",
      "include_in_plot": false,
      "score_10": 0.667,
      "date": "2025-09-29",
      "source": "Anthropic",
      "source_url": "https://www.anthropic.com/claude-sonnet-4-5-system-card",
      "trials": 30,
      "focus": "model"
    },
    {
      "agent": "Anthropic Agent",
      "model": "Claude Opus 4.1",
      "model_release_date": "2025-08-05",
      "include_in_plot": false,
      "score_10": 0.613,
      "date": "2025-09-29",
      "source": "Anthropic",
      "source_url": "https://www.anthropic.com/claude-sonnet-4-5-system-card",
      "trials": 30,
      "focus": "model"
    },
    {
      "agent": "Anthropic Agent",
      "model": "Claude Sonnet 4",
      "model_release_date": "2025-05-22",
      "include_in_plot": false,
      "score_10": 0.595,
      "date": "2025-09-29",
      "source": "Anthropic",
      "source_url": "https://www.anthropic.com/claude-sonnet-4-5-system-card",
      "trials": 30,
      "focus": "model"
    },
    {
      "agent": "Anthropic Agent",
      "model": "Claude Sonnet 3.7",
      "model_release_date": "2025-02-24",
      "include_in_plot": false,
      "score_10": 0.472,
      "date": "2025-09-29",
      "source": "Anthropic",
      "source_url": "https://www.anthropic.com/claude-sonnet-4-5-system-card",
      "trials": 30,
      "focus": "model"
    },
    {
      "agent": "Anthropic Agent",
      "model": "Claude Sonnet 4.5",
      "model_release_date": "2025-09-29",
      "include_in_plot": true,
      "score_10": 0.289,
      "date": "2025-09-29",
      "source": "Anthropic",
      "source_url": "https://www.anthropic.com/claude-sonnet-4-5-system-card",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Anthropic Agent",
      "model": "Claude Opus 4.1",
      "model_release_date": "2025-08-05",
      "include_in_plot": true,
      "score_10": 0.25,
      "date": "2025-09-29",
      "source": "Anthropic",
      "source_url": "https://www.anthropic.com/claude-sonnet-4-5-system-card",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Anthropic Agent",
      "model": "Claude Sonnet 4",
      "model_release_date": "2025-05-22",
      "include_in_plot": true,
      "score_10": 0.226,
      "date": "2025-09-29",
      "source": "Anthropic",
      "source_url": "https://www.anthropic.com/claude-sonnet-4-5-system-card",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Anthropic Agent",
      "model": "Claude Sonnet 3.7",
      "model_release_date": "2025-02-24",
      "include_in_plot": true,
      "score_10": 0.145,
      "date": "2025-09-29",
      "source": "Anthropic",
      "source_url": "https://www.anthropic.com/claude-sonnet-4-5-system-card",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenHands",
      "model": "Claude Sonnet 4",
      "model_release_date": "2025-05-22",
      "include_in_plot": false,
      "score_10": 0.17850033178500332,
      "score_x1": 0.019907100199071003,
      "date": "2025-05-23",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenHands",
      "model": "Claude Sonnet 3.7",
      "model_release_date": "2025-02-24",
      "include_in_plot": false,
      "score_10": 0.11944260119442601,
      "score_x1": 0.021897810218978103,
      "date": "2025-05-15",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenHands",
      "model": "GPT-5 (high)",
      "model_release_date": "2025-08-07",
      "include_in_plot": true,
      "score_10": 0.394,
      "date": "2025-12-05",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenHands",
      "model": "GPT-4.1",
      "model_release_date": "2025-04-14",
      "include_in_plot": true,
      "score_10": 0.09356337093563372,
      "score_x1": 0.012607830126078301,
      "date": "2025-05-15",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Cybench",
      "model": "GPT-4.1",
      "model_release_date": "2025-04-14",
      "include_in_plot": false,
      "score_10": 0.08958195089581951,
      "score_x1": 0.0225613802256138,
      "date": "2025-05-15",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "Codex CLI",
      "model": "GPT-4.1",
      "model_release_date": "2025-04-14",
      "include_in_plot": false,
      "score_10": 0.07365627073656271,
      "score_x1": 0.011944260119442602,
      "date": "2025-05-15",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "ENiGMA",
      "model": "GPT-4.1",
      "model_release_date": "2025-04-14",
      "include_in_plot": false,
      "score_10": 0.07232913072329131,
      "score_x1": 0.019243530192435302,
      "date": "2025-05-15",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenHands",
      "model": "Gemini 2.5 Flash Preview",
      "model_release_date": "2025-04-17",
      "include_in_plot": true,
      "score_10": 0.04844061048440611,
      "score_x1": 0.007962840079628402,
      "date": "2025-05-15",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenHands",
      "model": "DeepSeek-V3",
      "model_release_date": "2024-12-26",
      "include_in_plot": true,
      "score_10": 0.035832780358327807,
      "score_x1": 0.006635700066357001,
      "date": "2025-05-15",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenHands",
      "model": "o4-mini",
      "model_release_date": "2025-04-16",
      "include_in_plot": false,
      "score_10": 0.0245520902455209,
      "score_x1": 0.0006635700066357001,
      "date": "2025-05-15",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenHands",
      "model": "R2E-Gym-32B",
      "include_in_plot": false,
      "score_10": 0.019907100199071003,
      "score_x1": 0.005972130059721301,
      "date": "2025-05-15",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenHands",
      "model": "Qwen3-235B-A22B",
      "model_release_date": "2025-04-28",
      "include_in_plot": false,
      "score_10": 0.0185799601857996,
      "score_x1": 0.0033178500331785005,
      "date": "2025-05-15",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenHands",
      "model": "OpenHands-LM-32B",
      "include_in_plot": false,
      "score_10": 0.0165892501658925,
      "score_x1": 0.0033178500331785005,
      "date": "2025-05-15",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    },
    {
      "agent": "OpenHands",
      "model": "SWE-Gym-32B",
      "include_in_plot": false,
      "score_10": 0.0006635700066357001,
      "score_x1": 0.0006635700066357001,
      "date": "2025-05-15",
      "source": "CyberGym Team",
      "trials": 1,
      "focus": "model"
    }
  ],
  "level0": [
    {
      "agent": "OpenHands",
      "model": "GPT-4.1",
      "score_10": 0.034505640345056404,
      "score_x1": 0.0165892501658925,
      "date": "2025-05-15"
    }
  ],
  "level2": [
    {
      "agent": "OpenHands",
      "model": "GPT-4.1",
      "score_10": 0.13072329130723293,
      "score_x1": 0.0165892501658925,
      "date": "2025-05-15"
    }
  ],
  "level3": [
    {
      "agent": "OpenHands",
      "model": "GPT-4.1",
      "score_10": 0.1712010617120106,
      "score_x1": 0.013271400132714002,
      "date": "2025-05-15"
    }
  ]
}