date,arena,subset,category,provider,best_available_model,score,rank,ci_lower,ci_upper,gap_to_leader,leader_model,leader_org,n_models,raw_best_model,raw_best_score,replaced_retired,status,availability_note,availability_source,product
2026-10-22,text,text_style_control,overall,OpenAI,GPT-5.6 Sol (xHigh),1485.14,21,1480.88,1489.4,-32.03,Gemini 4 Argon (High),google,418,GPT-5.6 Sol (xHigh),1485.14,False,available,no deprecation listed (named as replacement model),https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-10-22,text,text_style_control,overall,Anthropic,Claude Opus 5.5 (High),1508.73,2,1501.94,1515.52,-8.44,Gemini 4 Argon (High),google,418,Claude Opus 5.5 (High),1508.73,False,available,Active,https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-10-22,text,text_style_control,overall,Google,Gemini 4 Argon (High),1517.17,1,1510.56,1523.79,0.0,Gemini 4 Argon (High),google,418,Gemini 4 Argon (High),1517.17,False,available,current lineup,https://ai.google.dev/gemini-api/docs/deprecations,Gemini
2026-10-22,text,text_style_control,coding,OpenAI,GPT-6 Astra (Max),1543.07,9,1532.09,1554.04,-8.5,Gemini 4 Argon (High),google,413,GPT-6 Astra (Max),1543.07,False,available,current lineup / no deprecation listed,https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-10-22,text,text_style_control,coding,Anthropic,Claude Fable 5 (High),1550.56,2,1544.07,1557.04,-1.01,Gemini 4 Argon (High),google,413,Claude Fable 5 (High),1550.56,False,available,"claude-fable-5 Active (retirement not before 2027-06-09); access suspended 2026-06-13..2026-07-01 under a US export-control order, redeployed 2026-07-01",https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-10-22,text,text_style_control,coding,Google,Gemini 4 Argon (High),1551.57,1,1539.03,1564.11,0.0,Gemini 4 Argon (High),google,413,Gemini 4 Argon (High),1551.57,False,available,current lineup,https://ai.google.dev/gemini-api/docs/deprecations,Gemini
2026-10-22,text,text_style_control,creative_writing,OpenAI,GPT-5.6 Sol (xHigh),1469.84,17,1462.36,1477.31,-49.76,Claude Opus 5.5 (High),anthropic,416,GPT-5.6 Sol (xHigh),1469.84,False,available,no deprecation listed (named as replacement model),https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-10-22,text,text_style_control,creative_writing,Anthropic,Claude Opus 5.5 (High),1519.6,1,1506.01,1533.2,0.0,Claude Opus 5.5 (High),anthropic,416,Claude Opus 5.5 (High),1519.6,False,available,Active,https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-10-22,text,text_style_control,creative_writing,Google,Gemini 4 Argon (High),1512.39,2,1498.88,1525.9,-7.21,Claude Opus 5.5 (High),anthropic,416,Gemini 4 Argon (High),1512.39,False,available,current lineup,https://ai.google.dev/gemini-api/docs/deprecations,Gemini
2026-10-22,text,text_style_control,math,OpenAI,GPT-6.1 Sol (Max),1513.37,8,1487.07,1539.67,-18.3,Gemini 4 Argon (High),google,404,GPT-6.1 Sol (Max),1513.37,False,available,current lineup / no deprecation listed,https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-10-22,text,text_style_control,math,Anthropic,Claude Fable 5.1 (Max),1528.98,2,1508.45,1549.51,-2.69,Gemini 4 Argon (High),google,404,Claude Fable 5.1 (Max),1528.98,False,available,Active,https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-10-22,text,text_style_control,math,Google,Gemini 4 Argon (High),1531.67,1,1507.51,1555.83,0.0,Gemini 4 Argon (High),google,404,Gemini 4 Argon (High),1531.67,False,available,current lineup,https://ai.google.dev/gemini-api/docs/deprecations,Gemini
2026-10-22,text,text_style_control,hard_prompts,OpenAI,GPT-5.6 Sol (xHigh),1511.45,17,1506.45,1516.45,-28.87,Gemini 4 Argon (High),google,418,GPT-5.6 Sol (xHigh),1511.45,False,available,no deprecation listed (named as replacement model),https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-10-22,text,text_style_control,hard_prompts,Anthropic,Claude Opus 5.5 (High),1534.62,2,1526.35,1542.9,-5.69,Gemini 4 Argon (High),google,418,Claude Opus 5.5 (High),1534.62,False,available,Active,https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-10-22,text,text_style_control,hard_prompts,Google,Gemini 4 Argon (High),1540.32,1,1532.14,1548.5,0.0,Gemini 4 Argon (High),google,418,Gemini 4 Argon (High),1540.32,False,available,current lineup,https://ai.google.dev/gemini-api/docs/deprecations,Gemini
2026-10-22,text,text_style_control,instruction_following,OpenAI,GPT-5.6 Sol (xHigh),1490.74,13,1484.69,1496.79,-30.65,Claude Opus 5.5 (High),anthropic,418,GPT-5.6 Sol (xHigh),1490.74,False,available,no deprecation listed (named as replacement model),https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-10-22,text,text_style_control,instruction_following,Anthropic,Claude Opus 5.5 (High),1521.39,1,1510.66,1532.12,0.0,Claude Opus 5.5 (High),anthropic,418,Claude Opus 5.5 (High),1521.39,False,available,Active,https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-10-22,text,text_style_control,instruction_following,Google,Gemini 4 Argon (High),1519.18,2,1508.7,1529.67,-2.2,Claude Opus 5.5 (High),anthropic,418,Gemini 4 Argon (High),1519.18,False,available,current lineup,https://ai.google.dev/gemini-api/docs/deprecations,Gemini
2026-10-22,text,text_style_control,longer_query,OpenAI,GPT-5.6 Sol (xHigh),1498.49,19,1492.8,1504.19,-36.01,Gemini 4 Argon (High),google,396,GPT-5.6 Sol (xHigh),1498.49,False,available,no deprecation listed (named as replacement model),https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-10-22,text,text_style_control,longer_query,Anthropic,Claude Opus 5.5 (High),1524.4,2,1514.34,1534.47,-10.09,Gemini 4 Argon (High),google,396,Claude Opus 5.5 (High),1524.4,False,available,Active,https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-10-22,text,text_style_control,longer_query,Google,Gemini 4 Argon (High),1534.5,1,1524.63,1544.36,0.0,Gemini 4 Argon (High),google,396,Gemini 4 Argon (High),1534.5,False,available,current lineup,https://ai.google.dev/gemini-api/docs/deprecations,Gemini
2026-10-22,text,text_style_control,multi_turn,OpenAI,GPT-5.4 (High),1492.94,21,1486.32,1499.56,-50.88,Gemini 4 Argon (High),google,416,GPT-5.2 Chat (20260210),1494.5,True,available,no deprecation listed for gpt-5.4,https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-10-22,text,text_style_control,multi_turn,Anthropic,Claude Fable 5 (High),1516.86,3,1509.07,1524.66,-26.96,Gemini 4 Argon (High),google,416,Claude Fable 5 (High),1516.86,False,available,"claude-fable-5 Active (retirement not before 2027-06-09); access suspended 2026-06-13..2026-07-01 under a US export-control order, redeployed 2026-07-01",https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-10-22,text,text_style_control,multi_turn,Google,Gemini 4 Argon (High),1543.82,1,1526.91,1560.73,0.0,Gemini 4 Argon (High),google,416,Gemini 4 Argon (High),1543.82,False,available,current lineup,https://ai.google.dev/gemini-api/docs/deprecations,Gemini
2026-10-22,text,text_style_control,expert,OpenAI,GPT-5.6 Sol (xHigh),1532.07,11,1522.85,1541.29,-17.7,Claude Fable 5 (High),anthropic,368,GPT-5.6 Sol (xHigh),1532.07,False,available,no deprecation listed (named as replacement model),https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-10-22,text,text_style_control,expert,Anthropic,Claude Fable 5 (High),1549.77,1,1540.62,1558.91,0.0,Claude Fable 5 (High),anthropic,368,Claude Fable 5 (High),1549.77,False,available,"claude-fable-5 Active (retirement not before 2027-06-09); access suspended 2026-06-13..2026-07-01 under a US export-control order, redeployed 2026-07-01",https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-10-22,text,text_style_control,expert,Google,Gemini 4 Argon (High),1537.04,5,1518.1,1555.98,-12.73,Claude Fable 5 (High),anthropic,368,Gemini 4 Argon (High),1537.04,False,available,current lineup,https://ai.google.dev/gemini-api/docs/deprecations,Gemini
2026-10-19,vision,vision_style_control,overall,OpenAI,GPT-6.1 Sol (Max),1293.53,9,1280.59,1306.47,-15.18,Claude Fable 5 (High),anthropic,167,GPT-6.1 Sol (Max),1293.53,False,available,current lineup / no deprecation listed,https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-10-19,vision,vision_style_control,overall,Anthropic,Claude Fable 5 (High),1308.71,1,1301.66,1315.77,0.0,Claude Fable 5 (High),anthropic,167,Claude Fable 5 (High),1308.71,False,available,"claude-fable-5 Active (retirement not before 2027-06-09); access suspended 2026-06-13..2026-07-01 under a US export-control order, redeployed 2026-07-01",https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-10-19,vision,vision_style_control,overall,Google,Gemini 3.8 Flash (High),1291.04,12,1281.23,1300.84,-17.68,Claude Fable 5 (High),anthropic,167,Gemini 3.7 Flash (High),1295.17,True,available,no shutdown date announced,https://ai.google.dev/gemini-api/docs/deprecations,Gemini
2026-10-23,webdev,webdev,overall,OpenAI,GPT-6 Astra (Max),1774.31,2,1764.4,1784.22,-32.61,Claude Opus 5.5 (Max),anthropic,150,GPT-6 Astra (Max),1774.31,False,available,current lineup / no deprecation listed,https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-10-23,webdev,webdev,overall,Anthropic,Claude Opus 5.5 (Max),1806.92,1,1793.56,1820.28,0.0,Claude Opus 5.5 (Max),anthropic,150,Claude Opus 5.5 (Max),1806.92,False,available,Active,https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-10-23,webdev,webdev,overall,Google,Gemini 4 Argon (High),1674.18,11,1661.54,1686.83,-132.74,Claude Opus 5.5 (Max),anthropic,150,Gemini 4 Argon (High),1674.18,False,available,current lineup,https://ai.google.dev/gemini-api/docs/deprecations,Gemini
2026-10-13,search,search_style_control,overall,OpenAI,gpt-5.6-sol-xhigh,1257.93,1,1251.46,1264.4,0.0,gpt-5.6-sol-xhigh,openai,42,gpt-5.6-sol-xhigh,1257.93,False,available,no deprecation listed (named as replacement model),https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-10-13,search,search_style_control,overall,Anthropic,claude-fable-5.1-high,1233.37,4,1226.2,1240.54,-24.56,gpt-5.6-sol-xhigh,openai,42,claude-fable-5.1-high,1233.37,False,available,Active,https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-10-13,search,search_style_control,overall,Google,gemini-3.1-pro-grounding,1206.62,14,1201.83,1211.4,-51.32,gpt-5.6-sol-xhigh,openai,42,gemini-3.1-pro-grounding,1206.62,False,available,gemini-3.1-pro-preview: no shutdown date announced (preview),https://ai.google.dev/gemini-api/docs/deprecations,Gemini
2026-09-13,document,document_style_control,overall,OpenAI,gpt-5.6-sol-xhigh,1488.67,8,1479.77,1497.57,-10.21,claude-fable-5,anthropic,44,gpt-5.6-sol-xhigh,1488.67,False,available,no deprecation listed (named as replacement model),https://developers.openai.com/api/docs/deprecations,ChatGPT
2026-09-13,document,document_style_control,overall,Anthropic,claude-fable-5,1498.88,1,1490.93,1506.82,0.0,claude-fable-5,anthropic,44,claude-fable-5,1498.88,False,available,"claude-fable-5 Active (retirement not before 2027-06-09); access suspended 2026-06-13..2026-07-01 under a US export-control order, redeployed 2026-07-01",https://docs.claude.com/en/docs/about-claude/model-deprecations,Claude
2026-09-13,document,document_style_control,overall,Google,gemini-3.1-pro-preview,1458.86,27,1453.56,1464.17,-40.01,claude-fable-5,anthropic,44,gemini-3.5-flash-medium,1460.56,True,available,gemini-3.1-pro-preview: no shutdown date announced (preview),https://ai.google.dev/gemini-api/docs/deprecations,Gemini