[
  {
    "benchmark": "swe-bench-multimodal",
    "score": 35.3,
    "metric": "solveable_accuracy",
    "cost_per_instance": 1.88,
    "average_runtime": 221.0,
    "full_archive": "https://results.eval.all-hands.dev/swebenchmultimodal/litellm_proxy-openai-gpt-5-5/1778132364/results.tar.gz",
    "tags": [
      "swe-bench-multimodal"
    ],
    "agent_version": "v1.19.1",
    "acp_agent_name": "codex-acp",
    "acp_agent_version": "v0.11.1",
    "submission_time": "2026-05-01T22:57:01+00:00",
    "component_scores": {
      "solveable_accuracy": 35.3,
      "unsolveable_accuracy": 0.0,
      "combined_accuracy": 23.5
    },
    "eval_visualization_page": "https://laminar.sh/shared/evals/7979f19b-67f3-4dff-b5a6-877398a8afe5"
  },
  {
    "benchmark": "swe-bench",
    "score": 78.0,
    "metric": "accuracy",
    "cost_per_instance": 1.17,
    "average_runtime": 171.0,
    "full_archive": "https://results.eval.all-hands.dev/swebench/litellm_proxy-openai-gpt-5-5/1778132371/results.tar.gz",
    "tags": [
      "swe-bench"
    ],
    "agent_version": "v1.19.1",
    "acp_agent_name": "codex-acp",
    "acp_agent_version": "v0.11.1",
    "submission_time": "2026-05-04T06:52:30+00:00",
    "eval_visualization_page": "https://laminar.sh/shared/evals/ee96327c-b067-4dd2-adc0-72204b08fbb3"
  },
  {
    "benchmark": "gaia",
    "score": 84.8,
    "metric": "accuracy",
    "cost_per_instance": 0.28,
    "average_runtime": 111.0,
    "full_archive": "https://results.eval.all-hands.dev/gaia/litellm_proxy-openai-gpt-5-5/1778132357/results.tar.gz",
    "tags": [
      "gaia"
    ],
    "agent_version": "v1.19.1",
    "acp_agent_name": "codex-acp",
    "acp_agent_version": "v0.11.1",
    "submission_time": "2026-05-04T03:11:56+00:00",
    "eval_visualization_page": "https://laminar.sh/shared/evals/d635f611-1e95-4934-a8b6-02eee6fb3fed"
  },
  {
    "benchmark": "swt-bench",
    "score": 80.4,
    "metric": "accuracy",
    "cost_per_instance": 1.11,
    "average_runtime": 146.0,
    "full_archive": "https://results.eval.all-hands.dev/swtbench/litellm_proxy-openai-gpt-5-5/1778132378/results.tar.gz",
    "tags": [
      "swt-bench"
    ],
    "agent_version": "v1.19.0",
    "acp_agent_name": "codex-acp",
    "acp_agent_version": "v0.11.1",
    "submission_time": "2026-04-30T23:18:25+00:00",
    "eval_visualization_page": "https://laminar.sh/shared/evals/d4ba8910-78eb-43bc-954b-2b5589d079e8"
  },
  {
    "benchmark": "commit0",
    "score": 37.5,
    "metric": "accuracy",
    "cost_per_instance": 5.57,
    "average_runtime": 425.0,
    "full_archive": "https://results.eval.all-hands.dev/commit0/litellm_proxy-openai-gpt-5-5/1778132350/results.tar.gz",
    "tags": [
      "commit0"
    ],
    "agent_version": "v1.19.1",
    "acp_agent_name": "codex-acp",
    "acp_agent_version": "v0.11.1",
    "submission_time": "2026-05-04T03:13:32+00:00",
    "eval_visualization_page": "https://laminar.sh/shared/evals/cef29694-8467-418d-bc74-859870453732"
  }
]