{"id":"gemini-4-argon","slug":"gemini-4-argon","name":"Gemini 4 Argon","orgId":"google","family":"gemini-4-argon","releaseDate":"2026-09-30","availability":"ga","openWeights":false,"license":"Proprietary","paramsTotal":null,"contextWindow":1048576,"knowledgeCutoff":null,"modelCardUrl":"https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/","hfRepo":null,"modalities":["text","audio","file","image","video"],"aliases":["gemini-4-argon-high","Gemini 4 Argon","gemini-4-argon_high","google/gemini-4-argon","Google: Gemini 4 Argon"],"aliasNote":null,"organization":{"id":"google","name":"Google DeepMind","slug":"google","country":null,"url":null,"color":"#4285F4","mark":"G"},"price":{"modelId":"gemini-4-argon","provider":"google","inputPerM":null,"outputPerM":null,"blendedPerM":4.000000000000001,"tokensPerS":null,"observedAt":"2026-10-13","sourceId":"openrouter"},"answer":{"word":"Yes.","scope":"SOTA in co-SOTA in Reasoning.","lead":"As of Oct 13, 2026, Gemini 4 Argon from Google DeepMind. SOTA in co-SOTA in Reasoning. Gemini 4 Argon is co-SOTA in Reasoning with low confidence because primary sources disagree beyond the margin of error (with GPT-6 Astra, Claude Opus 5.5).","confidence":"low"},"verdicts":[{"id":"v_overall_general-llm_gemini-4-argon_2026-10-13_open_PENDING","model":"gemini-4-argon","model_name":"Gemini 4 Argon","category":"general-llm","category_name":"General LLM","slice":"overall","state":"PENDING","rank":1,"margin":0,"confidence":"low","valid_from":"2026-10-13","valid_to":null,"configuration":"high","rationale":"Too new to call General LLM. The model is recent, and a crown change still needs two confirming signals.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pending · General LLM"},{"id":"v_pareto_general-llm_gemini-4-argon_2026-10-13_open_SOTA","model":"gemini-4-argon","model_name":"Gemini 4 Argon","category":"general-llm","category_name":"General LLM","slice":"pareto","state":"SOTA","rank":1,"margin":0,"confidence":"medium","valid_from":"2026-10-13","valid_to":null,"configuration":"high","rationale":"On the cost-quality frontier for General LLM: nothing scores better for less than $4.00 per 1M blended tokens.","safeguard_flags":["pareto"],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pareto · General LLM"},{"id":"v_overall_reasoning_gemini-4-argon_2026-10-08_open_CO_SOTA","model":"gemini-4-argon","model_name":"Gemini 4 Argon","category":"reasoning","category_name":"Reasoning","slice":"overall","state":"CO_SOTA","rank":1,"margin":0,"confidence":"low","valid_from":"2026-10-08","valid_to":null,"configuration":"high","rationale":"Gemini 4 Argon is co-SOTA in Reasoning with low confidence because primary sources disagree beyond the margin of error (with GPT-6 Astra, Claude Opus 5.5).","safeguard_flags":["source_conflict"],"co_sota_with":["gpt-6-astra","claude-opus-5-5"],"challengers":[],"methodology":"1.0","label":"Co-SOTA · Reasoning"},{"id":"v_pareto_reasoning_gemini-4-argon_2026-10-13_open_SOTA","model":"gemini-4-argon","model_name":"Gemini 4 Argon","category":"reasoning","category_name":"Reasoning","slice":"pareto","state":"SOTA","rank":1,"margin":0,"confidence":"medium","valid_from":"2026-10-13","valid_to":null,"configuration":"high","rationale":"On the cost-quality frontier for Reasoning: nothing scores better for less than $4.00 per 1M blended tokens.","safeguard_flags":["pareto"],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pareto · Reasoning"},{"id":"v_overall_science_gemini-4-argon_2026-10-13_open_PENDING","model":"gemini-4-argon","model_name":"Gemini 4 Argon","category":"science","category_name":"Science","slice":"overall","state":"PENDING","rank":1,"margin":null,"confidence":"low","valid_from":"2026-10-13","valid_to":null,"configuration":"high","rationale":"Too new to call Science. Last listed at #1 on 2026-09-30. That board does not award a crown.","safeguard_flags":["off_board"],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pending · Science"},{"id":"v_overall_coding_gemini-4-argon_2026-10-13_open_PENDING","model":"gemini-4-argon","model_name":"Gemini 4 Argon","category":"coding","category_name":"Coding","slice":"overall","state":"PENDING","rank":1,"margin":null,"confidence":"low","valid_from":"2026-10-13","valid_to":null,"configuration":"high","rationale":"Too new to call Coding. Ranked #1.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pending · Coding"},{"id":"v_overall_coding-web-dev_gemini-4-argon_2026-10-13_open_PENDING","model":"gemini-4-argon","model_name":"Gemini 4 Argon","category":"coding-web-dev","category_name":"Coding › Web dev","slice":"overall","state":"PENDING","rank":8,"margin":87.5,"confidence":"low","valid_from":"2026-10-13","valid_to":null,"configuration":"high","rationale":"Too new to call Coding › Web dev. The model is recent, and a crown change still needs two confirming signals.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pending · Coding › Web dev"},{"id":"v_pareto_coding-web-dev_gemini-4-argon_2026-10-13_open_NOT_SOTA","model":"gemini-4-argon","model_name":"Gemini 4 Argon","category":"coding-web-dev","category_name":"Coding › Web dev","slice":"pareto","state":"NOT_SOTA","rank":8,"margin":77.77777777777777,"confidence":"medium","valid_from":"2026-10-13","valid_to":null,"configuration":"high","rationale":"Not on the Coding › Web dev cost-quality frontier.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Not SOTA · #8"},{"id":"v_overall_long-context_gemini-4-argon_2026-10-13_open_PENDING","model":"gemini-4-argon","model_name":"Gemini 4 Argon","category":"long-context","category_name":"Long context","slice":"overall","state":"PENDING","rank":1,"margin":0,"confidence":"low","valid_from":"2026-10-13","valid_to":null,"configuration":"high","rationale":"Too new to call Long context. The model is recent, and a crown change still needs two confirming signals.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pending · Long context"},{"id":"v_pareto_long-context_gemini-4-argon_2026-10-13_open_NEAR_SOTA","model":"gemini-4-argon","model_name":"Gemini 4 Argon","category":"long-context","category_name":"Long context","slice":"pareto","state":"NEAR_SOTA","rank":1,"margin":0,"confidence":"medium","valid_from":"2026-10-13","valid_to":null,"configuration":"high","rationale":"Not on the Long context cost-quality frontier.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Near-SOTA · Long context"},{"id":"v_overall_factuality_gemini-4-argon_2026-10-13_open_PENDING","model":"gemini-4-argon","model_name":"Gemini 4 Argon","category":"factuality","category_name":"Factuality","slice":"overall","state":"PENDING","rank":1,"margin":0,"confidence":"low","valid_from":"2026-10-13","valid_to":null,"configuration":"high","rationale":"Too new to call Factuality. The model is recent, and a crown change still needs two confirming signals.","safeguard_flags":[],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pending · Factuality"},{"id":"v_pareto_factuality_gemini-4-argon_2026-10-13_open_SOTA","model":"gemini-4-argon","model_name":"Gemini 4 Argon","category":"factuality","category_name":"Factuality","slice":"pareto","state":"SOTA","rank":1,"margin":0,"confidence":"medium","valid_from":"2026-10-13","valid_to":null,"configuration":"high","rationale":"On the cost-quality frontier for Factuality: nothing scores better for less than $4.00 per 1M blended tokens.","safeguard_flags":["pareto"],"co_sota_with":[],"challengers":[],"methodology":"1.0","label":"Pareto · Factuality"}]}