{"record": {"aliases": ["FrontierChallenge", "Frontier Challenge"], "artifacts": [{"kind": "website", "url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/"}], "categories": ["scientific_agent"], "description": {"en": "Pass-rate results are model-scaffold measurements over the 97 released tasks with one trajectory per system-task pair, so a reported percentage is not a model-only number and does not estimate reliability beyond this release or under a different scaffold."}, "documents": [{"document_type": "benchmark_leaderboard", "id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "model_name": null, "organization": "ApodexAI", "published": null, "retrieved_at": "2026-09-02", "source": "model_reports", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "title": "FrontierChallenge leaderboard"}], "evidence_summary": {"document_count": 1, "model_count": 12, "model_count_basis": "source_model_id"}, "key": "model-reports:frontier_challenge", "modality": null, "name": "FrontierChallenge", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"registry": "data/model_cards.yml", "score_archive": "data/benchmark_scores.yml", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/"}, "publisher": null, "released": "2026-08-25", "released_reference": {"basis": "benchmark_release", "source_key": "model-reports:frontier_challenge", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/"}, "schema_version": 1, "sizes": [], "slug": "frontier_challenge", "source": "model_reports", "source_benchmark_id": "frontier_challenge"}, "schema_version": 1, "scores_by_source": {"model_reports": {"rows": [{"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "Apodex 1.1", "model_id": "[\"ApodexAI\",\"Apodex 1.1\"]", "model_name": "Apodex 1.1", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Apodex 1.1", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Apodex 1.1", "organization": "ApodexAI", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code", "rank_in_source_response": null, "raw_value": "10.3", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 10.3, "value_kind": "number"}, {"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "Claude Opus 5", "model_id": "[\"Anthropic\",\"Claude Opus 5\"]", "model_name": "Claude Opus 5", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Claude Opus 5", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Claude Opus 5", "organization": "Anthropic", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code", "rank_in_source_response": null, "raw_value": "17.5", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 17.5, "value_kind": "number"}, {"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "DeepSeek V4 Flash-0731", "model_id": "[\"DeepSeek\",\"DeepSeek V4 Flash-0731\"]", "model_name": "DeepSeek V4 Flash-0731", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000DeepSeek V4 Flash-0731", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000DeepSeek V4 Flash-0731", "organization": "DeepSeek", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code", "rank_in_source_response": null, "raw_value": "12.4", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 12.4, "value_kind": "number"}, {"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "DeepSeek V4 Pro-0813", "model_id": "[\"DeepSeek\",\"DeepSeek V4 Pro-0813\"]", "model_name": "DeepSeek V4 Pro-0813", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000DeepSeek V4 Pro-0813", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000DeepSeek V4 Pro-0813", "organization": "DeepSeek", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code", "rank_in_source_response": null, "raw_value": "13.4", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 13.4, "value_kind": "number"}, {"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "GLM-5.2", "model_id": "[\"Z.ai\",\"GLM-5.2\"]", "model_name": "GLM-5.2", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000GLM-5.2", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000GLM-5.2", "organization": "Z.ai", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code", "rank_in_source_response": null, "raw_value": "3.1", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 3.1, "value_kind": "number"}, {"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "Gemini 3.7 Flash", "model_id": "[\"Google\",\"Gemini 3.7 Flash\"]", "model_name": "Gemini 3.7 Flash", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Gemini 3.7 Flash", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Gemini 3.7 Flash", "organization": "Google", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code", "rank_in_source_response": null, "raw_value": "10.3", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 10.3, "value_kind": "number"}, {"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "Grok 4.6", "model_id": "[\"xAI\",\"Grok 4.6\"]", "model_name": "Grok 4.6", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Grok 4.6", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Grok 4.6", "organization": "xAI", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code", "rank_in_source_response": null, "raw_value": "20.6", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 20.6, "value_kind": "number"}, {"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "Kimi K3", "model_id": "[\"Moonshot AI\",\"Kimi K3\"]", "model_name": "Kimi K3", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Kimi K3", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Kimi K3", "organization": "Moonshot AI", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code", "rank_in_source_response": null, "raw_value": "17.5", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 17.5, "value_kind": "number"}, {"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "Qwen 3.8 Max", "model_id": "[\"Qwen\",\"Qwen 3.8 Max\"]", "model_name": "Qwen 3.8 Max", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Qwen 3.8 Max", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Qwen 3.8 Max", "organization": "Qwen", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code", "rank_in_source_response": null, "raw_value": "15.5", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 15.5, "value_kind": "number"}, {"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "Qwen3.5-397B-A17B", "model_id": "[\"Qwen\",\"Qwen3.5-397B-A17B\"]", "model_name": "Qwen3.5-397B-A17B", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Qwen3.5-397B-A17B", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Qwen3.5-397B-A17B", "organization": "Qwen", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code", "rank_in_source_response": null, "raw_value": "4.1", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 4.1, "value_kind": "number"}, {"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Codex\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "GPT-5.6 Sol", "model_id": "[\"OpenAI\",\"GPT-5.6 Sol\"]", "model_name": "GPT-5.6 Sol", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Codex\u0000GPT-5.6 Sol", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Codex\u0000GPT-5.6 Sol", "organization": "OpenAI", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Codex", "rank_in_source_response": null, "raw_value": "20.6", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 20.6, "value_kind": "number"}, {"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Codex\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "GPT-5.6 Terra (max)", "model_id": "[\"OpenAI\",\"GPT-5.6 Terra (max)\"]", "model_name": "GPT-5.6 Terra (max)", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Codex\u0000GPT-5.6 Terra (max)", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Codex\u0000GPT-5.6 Terra (max)", "organization": "OpenAI", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Codex", "rank_in_source_response": null, "raw_value": "15.5", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 15.5, "value_kind": "number"}, {"benchmark_id": "frontier_challenge", "comparable_group": "[\"model-reports:frontier_challenge\",\"frontier_challenge_97_tasks_2026_08_25\",\"97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Frontier Agent (Agent Team)\"]", "date_precision": "document_publication", "document_id": "model_reports:frontier_challenge_leaderboard_2026_08_25", "instrument": "frontier_challenge_97_tasks_2026_08_25", "key": "model-reports:frontier_challenge", "measured_by": "ApodexAI", "measurement_kind": "benchmark_publisher_run", "model": "Apodex 1.1", "model_id": "[\"ApodexAI\",\"Apodex 1.1\"]", "model_name": "Apodex 1.1", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Frontier Agent (Agent Team)\u0000Apodex 1.1", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Frontier Agent (Agent Team)\u0000Apodex 1.1", "organization": "ApodexAI", "protocol": "97 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Frontier Agent (Agent Team)", "rank_in_source_response": null, "raw_value": "12.4", "read_from": "html_text", "reported_at": "2026-08-25", "reported_by": "third_party", "reported_date": "2026-08-25", "series_id": "model_reports:frontier_challenge:default", "source": "model_reports", "source_document_type": "benchmark_leaderboard", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_title": "FrontierChallenge leaderboard", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/", "value": 12.4, "value_kind": "number"}], "series": {"declared_max": null, "direction": "higher_is_better", "direction_basis": "reported_metric", "display_scale": null, "first_score_record": {"date_precision": "document_publication", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Apodex 1.1", "reported_at": "2026-08-25", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/"}, "first_score_report": {"date_precision": "document_publication", "obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Apodex 1.1", "reported_at": "2026-08-25", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/"}, "key": "model-reports:frontier_challenge", "max_score_contradicted": false, "metric": "pass_rate", "observation_count": 13, "observed_max": 20.6, "score_summary": {"display_max": 20.6, "display_multiplier": 1, "model_count": 12, "model_count_basis": "source_model_id", "numeric_count": 13, "raw_max": 20.6, "raw_min": 3.1, "source_reference": {"obs_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Grok 4.6", "observation_id": "curated\u0000frontier_challenge\u0000frontier_challenge_leaderboard_2026_08_25\u0000frontier_challenge_97_tasks_2026_08_25\u000097 released tasks; one trajectory per system-task pair; pass = native Score >= 99.9; agent scaffold: Claude Code\u0000Grok 4.6", "reported_at": "2026-08-25", "reported_date": "2026-08-25", "source_id": "frontier_challenge_leaderboard_2026_08_25", "source_url": "https://apodexai.github.io/FrontierAgent/benchmarks/FrontierChallenge/"}, "unit": "percent"}, "series_id": "model_reports:frontier_challenge:default", "unit": "percent"}}}, "siblings": []}