BenchAtlas

Rankings / OpenAI

GPT 4 1106 Preview

released 2023-11-06

BenchAtlas Index

as of 2026-09-11
Not enough benchmark coverage for a composite score (needs ≥4 families across ≥3 categories). The per-benchmark evidence below stands on its own — sparse coverage is an evidence gap, not a low score.

Benchmark evidence

31 results
external indices
ECI126.5 pointsindependentEpoch AI Benchmarking Hub
human preference
Coding (style control)older version
arena=text · category=coding · style_control=true
1339.9
1332.5–1347.4
communityLMArena Leaderboard Dataset
French (style control)older version
arena=text · category=french · style_control=true
1339.6
1318.2–1361.0
communityLMArena Leaderboard Dataset
English (style control)older version
arena=text · category=english · style_control=true
1325.1
1320.2–1330.0
communityLMArena Leaderboard Dataset
Hard Prompts English (style control)older version
arena=text · category=hard_prompts_english · style_control=true
1324.1
1317.0–1331.3
communityLMArena Leaderboard Dataset
Industry Software And It Services (style control)older version
arena=text · category=industry_software_and_it_services · style_control=true
1324.1
1317.8–1330.4
communityLMArena Leaderboard Dataset
Industry Life And Physical And Social Science (style control)older version
arena=text · category=industry_life_and_physical_and_social_science · style_control=true
1319.1
1311.4–1326.8
communityLMArena Leaderboard Dataset
Chinese (style control)older version
arena=text · category=chinese · style_control=true
1314.1
1304.0–1324.2
communityLMArena Leaderboard Dataset
Industry Legal And Government (style control)older version
arena=text · category=industry_legal_and_government · style_control=true
1314.0
1302.6–1325.5
communityLMArena Leaderboard Dataset
Overall (style control)
arena=text · category=overall · style_control=true
1312.4
1308.5–1316.3
communityLMArena Leaderboard Dataset
Hard Prompts (style control)older version
arena=text · category=hard_prompts · style_control=true
1312.0
1306.0–1318.1
communityLMArena Leaderboard Dataset
Spanish (style control)older version
arena=text · category=spanish · style_control=true
1311.2
1289.0–1333.4
communityLMArena Leaderboard Dataset
Industry Mathematical (style control)older version
arena=text · category=industry_mathematical · style_control=true
1308.6
1300.4–1316.8
communityLMArena Leaderboard Dataset
Industry Writing And Literature And Language (style control)older version
arena=text · category=industry_writing_and_literature_and_language · style_control=true
1306.7
1300.2–1313.2
communityLMArena Leaderboard Dataset
Math (style control)older version
arena=text · category=math · style_control=true
1302.8
1294.9–1310.6
communityLMArena Leaderboard Dataset
Industry Medicine And Healthcare (style control)older version
arena=text · category=industry_medicine_and_healthcare · style_control=true
1302.7
1290.7–1314.6
communityLMArena Leaderboard Dataset
Longer Query (style control)older version
arena=text · category=longer_query · style_control=true
1301.8
1293.3–1310.3
communityLMArena Leaderboard Dataset
Russian (style control)older version
arena=text · category=russian · style_control=true
1299.7
1290.4–1308.9
communityLMArena Leaderboard Dataset
Multi Turn (style control)older version
arena=text · category=multi_turn · style_control=true
1298.5
1290.7–1306.3
communityLMArena Leaderboard Dataset
German (style control)older version
arena=text · category=german · style_control=true
1297.8
1282.4–1313.3
communityLMArena Leaderboard Dataset
Instruction Following (style control)older version
arena=text · category=instruction_following · style_control=true
1296.0
1290.4–1301.7
communityLMArena Leaderboard Dataset
Non English (style control)older version
arena=text · category=non_english · style_control=true
1295.3
1289.8–1300.7
communityLMArena Leaderboard Dataset
Industry Entertainment And Sports And Media (style control)older version
arena=text · category=industry_entertainment_and_sports_and_media · style_control=true
1289.8
1282.1–1297.5
communityLMArena Leaderboard Dataset
Expert (style control)older version
arena=text · category=expert · style_control=true
1289.8
1277.7–1301.8
communityLMArena Leaderboard Dataset
Creative Writing (style control)older version
arena=text · category=creative_writing · style_control=true
1285.6
1277.7–1293.6
communityLMArena Leaderboard Dataset
Industry Business And Management And Financial Operations (style control)older version
arena=text · category=industry_business_and_management_and_financial_operations · style_control=true
1278.9
1270.1–1287.7
communityLMArena Leaderboard Dataset
Japanese (style control)older version
arena=text · category=japanese · style_control=true
1262.2
1243.7–1280.7
communityLMArena Leaderboard Dataset
Korean (style control)older version
arena=text · category=korean · style_control=true
1252.4
1231.0–1273.8
communityLMArena Leaderboard Dataset
Exclude Ties (style control)older version
arena=text · category=exclude_ties · style_control=true
1244.6
1238.5–1250.6
communityLMArena Leaderboard Dataset
knowledge science
GPQA Diamond42.4 %independentEpoch AI Benchmarking Hub
2025-01-27
reasoning math
MATH Level 540.0 %independentEpoch AI Benchmarking Hub
2025-01-27

Agent + model results

systems, not bare-model scores
agent + model CodeR + GPT 4 1106 PreviewSWE-bench Lite28.3 %unverifiedSWE-bench Leaderboard
agent + model SWE-agent + GPT 4 1106 PreviewSWE-bench Verified22.4 %communitySWE-bench Leaderboard
agent + model SWE-agent + GPT 4 1106 PreviewSWE-bench Lite18.0 %communitySWE-bench Leaderboard
agent + model SWE-agent + GPT 4 1106 PreviewSWE-bench (full test split)12.5 %communitySWE-bench Leaderboard
agent + model RAG + GPT 4 1106 PreviewSWE-bench Verified2.8 %communitySWE-bench Leaderboard
agent + model RAG + GPT 4 1106 PreviewSWE-bench Lite2.7 %communitySWE-bench Leaderboard
agent + model RAG + GPT 4 1106 PreviewSWE-bench (full test split)1.3 %communitySWE-bench Leaderboard

These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.