BenchAtlas

Rankings / Google

Gemini 3 Flash

proprietary

release date unknown

BenchAtlas Index

as of 2026-07-14
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

57 results
human preference
Chinese (style control)older version
arena=text · category=chinese · style_control=true
1521.4
1506.01536.7
communityLMArena Leaderboard Dataset
Coding (style control)older version
arena=text · category=coding · style_control=true
1508.5
1500.51516.4
communityLMArena Leaderboard Dataset
French (style control)older version
arena=text · category=french · style_control=true
1505.2
1482.21528.1
communityLMArena Leaderboard Dataset
Industry Software And It Services (style control)older version
arena=text · category=industry_software_and_it_services · style_control=true
1500.8
1494.21507.4
communityLMArena Leaderboard Dataset
Polish (style control)older version
arena=text · category=polish · style_control=true
1497.0
1477.01517.1
communityLMArena Leaderboard Dataset
Expert (style control)older version
arena=text · category=expert · style_control=true
1496.9
1483.61510.2
communityLMArena Leaderboard Dataset
German (style control)older version
arena=text · category=german · style_control=true
1495.1
1468.71521.4
communityLMArena Leaderboard Dataset
Chinese (style control)older version
arena=text · category=chinese · style_control=true
minimal effort1494.8
1484.21505.3
communityLMArena Leaderboard Dataset
Hard Prompts (style control)older version
arena=text · category=hard_prompts · style_control=true
1492.4
1486.81498.0
communityLMArena Leaderboard Dataset
Coding (style control)older version
arena=text · category=coding · style_control=true
minimal effort1492.0
1486.71497.2
communityLMArena Leaderboard Dataset
Hard Prompts English (style control)older version
arena=text · category=hard_prompts_english · style_control=true
1487.6
1480.21494.9
communityLMArena Leaderboard Dataset
Industry Legal And Government (style control)older version
arena=text · category=industry_legal_and_government · style_control=true
1487.0
1474.51499.6
communityLMArena Leaderboard Dataset
Industry Software And It Services (style control)older version
arena=text · category=industry_software_and_it_services · style_control=true
minimal effort1486.9
1482.31491.5
communityLMArena Leaderboard Dataset
Russian (style control)older version
arena=text · category=russian · style_control=true
1486.0
1475.61496.5
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
1486.0
1477.21494.9
communityLMArena Leaderboard Dataset
Exclude Ties (style control)older version
arena=text · category=exclude_ties · style_control=true
1485.4
1479.11491.7
communityLMArena Leaderboard Dataset
Industry Medicine And Healthcare (style control)older version
arena=text · category=industry_medicine_and_healthcare · style_control=true
1485.3
1472.11498.6
communityLMArena Leaderboard Dataset
Multi Turn (style control)older version
arena=text · category=multi_turn · style_control=true
1483.4
1474.81492.1
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
minimal effort1481.5
1475.51487.5
communityLMArena Leaderboard Dataset
French (style control)older version
arena=text · category=french · style_control=true
minimal effort1479.3
1464.81493.9
communityLMArena Leaderboard Dataset
Polish (style control)older version
arena=text · category=polish · style_control=true
minimal effort1479.1
1464.71493.5
communityLMArena Leaderboard Dataset
Industry Medicine And Healthcare (style control)older version
arena=text · category=industry_medicine_and_healthcare · style_control=true
minimal effort1478.1
1469.51486.7
communityLMArena Leaderboard Dataset
German (style control)older version
arena=text · category=german · style_control=true
minimal effort1477.8
1459.81495.9
communityLMArena Leaderboard Dataset
Multi Turn (style control)older version
arena=text · category=multi_turn · style_control=true
minimal effort1477.0
1471.11483.0
communityLMArena Leaderboard Dataset
Hard Prompts (style control)older version
arena=text · category=hard_prompts · style_control=true
minimal effort1476.0
1472.01480.0
communityLMArena Leaderboard Dataset
Math (style control)older version
arena=text · category=math · style_control=true
1475.9
1462.81489.0
communityLMArena Leaderboard Dataset
Longer Query (style control)older version
arena=text · category=longer_query · style_control=true
1475.5
1468.21482.9
communityLMArena Leaderboard Dataset
English (style control)older version
arena=text · category=english · style_control=true
1474.9
1468.81480.9
communityLMArena Leaderboard Dataset
Overall (style control)
arena=text · category=overall · style_control=true
1473.2
1468.81477.6
communityLMArena Leaderboard Dataset
Hard Prompts English (style control)older version
arena=text · category=hard_prompts_english · style_control=true
minimal effort1473.0
1468.01477.9
communityLMArena Leaderboard Dataset
Industry Mathematical (style control)older version
arena=text · category=industry_mathematical · style_control=true
1472.3
1456.71487.9
communityLMArena Leaderboard Dataset
Industry Legal And Government (style control)older version
arena=text · category=industry_legal_and_government · style_control=true
minimal effort1471.3
1463.21479.5
communityLMArena Leaderboard Dataset
Russian (style control)older version
arena=text · category=russian · style_control=true
minimal effort1470.2
1463.01477.4
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
1470.0
1461.71478.3
communityLMArena Leaderboard Dataset
Spanish (style control)older version
arena=text · category=spanish · style_control=true
1469.7
1448.81490.7
communityLMArena Leaderboard Dataset
Non English (style control)older version
arena=text · category=non_english · style_control=true
1467.4
1461.71473.0
communityLMArena Leaderboard Dataset
Expert (style control)older version
arena=text · category=expert · style_control=true
minimal effort1465.8
1457.81473.8
communityLMArena Leaderboard Dataset
Exclude Ties (style control)older version
arena=text · category=exclude_ties · style_control=true
minimal effort1464.6
1460.21469.0
communityLMArena Leaderboard Dataset
Longer Query (style control)older version
arena=text · category=longer_query · style_control=true
minimal effort1463.1
1458.21468.0
communityLMArena Leaderboard Dataset
English (style control)older version
arena=text · category=english · style_control=true
minimal effort1462.3
1458.11466.5
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
1460.9
1453.01468.9
communityLMArena Leaderboard Dataset
Overall (style control)
arena=text · category=overall · style_control=true
minimal effort1459.6
1456.41462.8
communityLMArena Leaderboard Dataset
Instruction Following (style control)older version
arena=text · category=instruction_following · style_control=true
1458.4
1451.11465.6
communityLMArena Leaderboard Dataset
Creative Writing (style control)older version
arena=text · category=creative_writing · style_control=true
1458.3
1448.81467.8
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
minimal effort1455.6
1450.01461.3
communityLMArena Leaderboard Dataset
Math (style control)older version
arena=text · category=math · style_control=true
minimal effort1454.0
1444.81463.3
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
1453.3
1444.71461.8
communityLMArena Leaderboard Dataset
Non English (style control)older version
arena=text · category=non_english · style_control=true
minimal effort1452.3
1448.21456.4
communityLMArena Leaderboard Dataset
Industry Mathematical (style control)older version
arena=text · category=industry_mathematical · style_control=true
minimal effort1450.1
1440.01460.2
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
minimal effort1448.7
1443.31454.1
communityLMArena Leaderboard Dataset
Creative Writing (style control)older version
arena=text · category=creative_writing · style_control=true
minimal effort1447.9
1441.71454.2
communityLMArena Leaderboard Dataset
Korean (style control)older version
arena=text · category=korean · style_control=true
1447.0
1420.81473.3
communityLMArena Leaderboard Dataset
Spanish (style control)older version
arena=text · category=spanish · style_control=true
minimal effort1445.1
1430.91459.3
communityLMArena Leaderboard Dataset
Instruction Following (style control)older version
arena=text · category=instruction_following · style_control=true
minimal effort1444.1
1439.21449.0
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
minimal effort1440.7
1435.01446.4
communityLMArena Leaderboard Dataset
Japanese (style control)older version
arena=text · category=japanese · style_control=true
minimal effort1439.7
1414.91464.6
communityLMArena Leaderboard Dataset
Korean (style control)older version
arena=text · category=korean · style_control=true
minimal effort1433.1
1414.41451.9
communityLMArena Leaderboard Dataset

Agent + model results

systems, not bare-model scores
agent + model mini-SWE-agent + Gemini 3 FlashSWE-bench Multilingual72.7 %communitySWE-bench Leaderboard
agent + model mini-SWE-agent + Gemini 3 FlashSWE-bench Multilingual0.4 usd_per_taskcommunitySWE-bench Leaderboard

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