{"id":"gpt-6-luna","slug":"gpt-6-luna","name":"GPT-6 Luna","orgId":"openai","family":"gpt-6-luna","releaseDate":"2026-09-22","availability":"ga","openWeights":false,"license":"Proprietary","paramsTotal":null,"contextWindow":1050000,"knowledgeCutoff":null,"modelCardUrl":"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset","hfRepo":null,"modalities":["text","file","image"],"aliases":["gpt-6-luna-max","GPT-6 Luna","gpt-6-luna_max","gpt-6-luna_xhigh","gpt-6-luna_high","gpt-6-luna_medium","gpt-6-luna_low","gpt-6-luna_none","GPT-6 Luna max","openai/gpt-6-luna","OpenAI: GPT-6 Luna"],"aliasNote":null,"organization":{"id":"openai","name":"OpenAI","slug":"openai","country":null,"url":null,"color":"#10A37F","mark":"O"},"price":{"modelId":"gpt-6-luna","provider":"openai","inputPerM":null,"outputPerM":null,"blendedPerM":0.19999999999999998,"tokensPerS":null,"observedAt":"2026-10-14","sourceId":"openrouter"},"answer":{"word":"Yes.","scope":"Yes on Coding (best per dollar); Coding › Agentic (best per dollar). Not the overall leader.","lead":"As of Oct 14, 2026, GPT-6 Luna from OpenAI. Yes on Coding (best per dollar); Coding › Agentic (best per dollar). Not the overall leader. Ranked #17 in Coding.","confidence":"medium"},"verdicts":[{"id":"v_overall_general-llm_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"general-llm","category_name":"General LLM","slice":"overall","state":"RANKED","rank":65,"present_evidence":[],"missing_evidence":[],"margin":44.13793103448276,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Ranked #65 in General LLM.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#65 in General LLM"},{"id":"v_pareto_general-llm_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"general-llm","category_name":"General LLM","slice":"pareto","state":"RANKED","rank":82,"present_evidence":[],"missing_evidence":[],"margin":43.61702127659574,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Not on the General LLM cost-quality frontier.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#82 in General LLM"},{"id":"v_overall_reasoning_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"reasoning","category_name":"Reasoning","slice":"overall","state":"RANKED","rank":61,"present_evidence":[],"missing_evidence":[],"margin":42.361111111111114,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Ranked #61 in Reasoning.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#61 in Reasoning"},{"id":"v_pareto_reasoning_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"reasoning","category_name":"Reasoning","slice":"pareto","state":"RANKED","rank":76,"present_evidence":[],"missing_evidence":[],"margin":40.95744680851064,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Not on the Reasoning cost-quality frontier.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#76 in Reasoning"},{"id":"v_overall_math_gpt-6-luna_2026-10-14_open_INSUFFICIENT_DATA","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"math","category_name":"Math","slice":"overall","state":"INSUFFICIENT_DATA","rank":null,"present_evidence":[{"benchmark":"lmarena-text-math","name":"LMArena Text (style-controlled, math)","rank":60}],"missing_evidence":[{"benchmark":"hle","name":"Humanity's Last Exam"},{"benchmark":"frontiermath","name":"FrontierMath"}],"margin":null,"confidence":"low","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"#60 on LMArena Text (style-controlled, math); missing Humanity's Last Exam and FrontierMath","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Not enough data"},{"id":"v_overall_science_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"science","category_name":"Science","slice":"overall","state":"RANKED","rank":33,"present_evidence":[],"missing_evidence":[],"margin":null,"confidence":"low","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Last listed at #33 in Science on 2026-09-22. That board does not award a crown.","safeguard_flags":["off_board"],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#33 in Science"},{"id":"v_overall_coding_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"coding","category_name":"Coding","slice":"overall","state":"RANKED","rank":17,"present_evidence":[],"missing_evidence":[],"margin":81.81818181818181,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Ranked #17 in Coding.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#17 in Coding"},{"id":"v_pareto_coding_gpt-6-luna_2026-10-14_open_SOTA","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"coding","category_name":"Coding","slice":"pareto","state":"SOTA","rank":17,"present_evidence":[],"missing_evidence":[],"margin":81.81818181818181,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"On the cost-quality frontier for Coding: nothing scores better for less than $0.20 per 1M blended tokens.","safeguard_flags":["pareto"],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pareto · Coding"},{"id":"v_overall_coding-agentic_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"coding-agentic","category_name":"Coding › Agentic","slice":"overall","state":"RANKED","rank":17,"present_evidence":[],"missing_evidence":[],"margin":81.81818181818181,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Ranked #17 in Coding › Agentic.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#17 in Coding › Agentic"},{"id":"v_pareto_coding-agentic_gpt-6-luna_2026-10-14_open_SOTA","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"coding-agentic","category_name":"Coding › Agentic","slice":"pareto","state":"SOTA","rank":17,"present_evidence":[],"missing_evidence":[],"margin":81.81818181818181,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"On the cost-quality frontier for Coding › Agentic: nothing scores better for less than $0.20 per 1M blended tokens.","safeguard_flags":["pareto"],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pareto · Coding › Agentic"},{"id":"v_overall_coding-web-dev_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"coding-web-dev","category_name":"Coding › Web dev","slice":"overall","state":"RANKED","rank":28,"present_evidence":[],"missing_evidence":[],"margin":25.714285714285708,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Ranked #28 in Coding › Web dev.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#28 in Coding › Web dev"},{"id":"v_pareto_coding-web-dev_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"coding-web-dev","category_name":"Coding › Web dev","slice":"pareto","state":"RANKED","rank":31,"present_evidence":[],"missing_evidence":[],"margin":24,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Not on the Coding › Web dev cost-quality frontier.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#31 in Coding › Web dev"},{"id":"v_overall_long-context_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"long-context","category_name":"Long context","slice":"overall","state":"RANKED","rank":60,"present_evidence":[],"missing_evidence":[],"margin":41.379310344827594,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Ranked #60 in Long context.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#60 in Long context"},{"id":"v_pareto_long-context_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"long-context","category_name":"Long context","slice":"pareto","state":"RANKED","rank":75,"present_evidence":[],"missing_evidence":[],"margin":39.89361702127659,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Not on the Long context cost-quality frontier.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#75 in Long context"},{"id":"v_overall_vision_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"vision","category_name":"Vision","slice":"overall","state":"RANKED","rank":17,"present_evidence":[],"missing_evidence":[],"margin":47.36842105263158,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Ranked #17 in Vision.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#17 in Vision"},{"id":"v_pareto_vision_gpt-6-luna_2026-10-14_open_SOTA","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"vision","category_name":"Vision","slice":"pareto","state":"SOTA","rank":17,"present_evidence":[],"missing_evidence":[],"margin":47.36842105263158,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"On the cost-quality frontier for Vision: nothing scores better for less than $0.20 per 1M blended tokens.","safeguard_flags":["pareto"],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pareto · Vision"},{"id":"v_overall_factuality_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"factuality","category_name":"Factuality","slice":"overall","state":"RANKED","rank":39,"present_evidence":[],"missing_evidence":[],"margin":29.545454545454547,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Ranked #39 in Factuality.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#39 in Factuality"},{"id":"v_pareto_factuality_gpt-6-luna_2026-10-14_open_SOTA","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"factuality","category_name":"Factuality","slice":"pareto","state":"SOTA","rank":45,"present_evidence":[],"missing_evidence":[],"margin":26.74418604651163,"confidence":"medium","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"On the cost-quality frontier for Factuality: nothing scores better for less than $0.20 per 1M blended tokens.","safeguard_flags":["pareto"],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pareto · Factuality"},{"id":"v_overall_arc_gpt-6-luna_2026-10-14_open_RANKED","model":"gpt-6-luna","model_name":"GPT-6 Luna","category":"arc","category_name":"Abstraction (ARC)","slice":"overall","state":"RANKED","rank":22,"present_evidence":[],"missing_evidence":[],"margin":null,"confidence":"low","valid_from":"2026-10-14","valid_to":null,"configuration":"max","rationale":"Last listed at #22 in Abstraction (ARC) on 2026-09-22. That board does not award a crown.","safeguard_flags":["off_board"],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"#22 in Abstraction (ARC)"}]}