Rankings / Google DeepMind
Gemini 2.5 Flash Preview 09 2025
released 2025-09-25
BenchAtlas Index
as of 2026-09-11
Benchmark evidence
46 results
external indices
| ECI | 16K budget | 143.0 points | independent | Epoch AI Benchmarking Hub |
| ECI | 143.0 points | independent | Epoch AI Benchmarking Hub |
human preference
| Chinese (style control)older version arena=text · category=chinese · style_control=true | 1441.3 1428.8–1453.8 | community | LMArena 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 | 1434.9 1426.4–1443.4 | community | LMArena Leaderboard Dataset | |
| French (style control)older version arena=text · category=french · style_control=true | 1434.7 1409.4–1460.0 | community | LMArena Leaderboard Dataset | |
| Expert (style control)older version arena=text · category=expert · style_control=true | 1431.8 1417.4–1446.2 | community | LMArena Leaderboard Dataset | |
| Coding (style control)older version arena=text · category=coding · style_control=true | 1428.4 1420.9–1435.9 | community | LMArena Leaderboard Dataset | |
| Industry Medicine And Healthcare (style control)older version arena=text · category=industry_medicine_and_healthcare · style_control=true | 1427.8 1412.7–1442.8 | community | LMArena Leaderboard Dataset | |
| Industry Legal And Government (style control)older version arena=text · category=industry_legal_and_government · style_control=true | 1426.6 1413.8–1439.5 | community | LMArena Leaderboard Dataset | |
| Hard Prompts English (style control)older version arena=text · category=hard_prompts_english · style_control=true | 1425.3 1418.0–1432.5 | community | LMArena Leaderboard Dataset | |
| Industry Software And It Services (style control)older version arena=text · category=industry_software_and_it_services · style_control=true | 1422.6 1416.5–1428.7 | community | LMArena Leaderboard Dataset | |
| Hard Prompts (style control)older version arena=text · category=hard_prompts · style_control=true | 1420.1 1414.9–1425.3 | community | LMArena Leaderboard Dataset | |
| Longer Query (style control)older version arena=text · category=longer_query · style_control=true | 1416.8 1409.7–1423.9 | community | LMArena Leaderboard Dataset | |
| Industry Mathematical (style control)older version arena=text · category=industry_mathematical · style_control=true | 1416.5 1402.0–1430.9 | community | LMArena Leaderboard Dataset | |
| German (style control)older version arena=text · category=german · style_control=true | 1413.8 1390.3–1437.3 | community | LMArena Leaderboard Dataset | |
| English (style control)older version arena=text · category=english · style_control=true | 1412.8 1407.1–1418.4 | community | LMArena Leaderboard Dataset | |
| Math (style control)older version arena=text · category=math · style_control=true | 1409.5 1396.4–1422.7 | community | LMArena Leaderboard Dataset | |
| Overall (style control) arena=text · category=overall · style_control=true | 1403.5 1399.5–1407.6 | community | LMArena 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 | 1403.4 1395.4–1411.4 | community | LMArena Leaderboard Dataset | |
| Instruction Following (style control)older version arena=text · category=instruction_following · style_control=true | 1400.8 1394.2–1407.5 | community | LMArena Leaderboard Dataset | |
| Polish (style control)older version arena=text · category=polish · style_control=true | 1400.4 1383.5–1417.3 | community | LMArena Leaderboard Dataset | |
| Multi Turn (style control)older version arena=text · category=multi_turn · style_control=true | 1399.9 1391.7–1408.0 | community | LMArena Leaderboard Dataset | |
| Russian (style control)older version arena=text · category=russian · style_control=true | 1394.7 1383.9–1405.4 | community | LMArena Leaderboard Dataset | |
| Industry Writing And Literature And Language (style control)older version arena=text · category=industry_writing_and_literature_and_language · style_control=true | 1394.3 1386.9–1401.8 | community | LMArena Leaderboard Dataset | |
| Spanish (style control)older version arena=text · category=spanish · style_control=true | 1391.7 1371.6–1411.8 | community | LMArena Leaderboard Dataset | |
| Non English (style control)older version arena=text · category=non_english · style_control=true | 1390.8 1385.7–1395.9 | community | LMArena Leaderboard Dataset | |
| Exclude Ties (style control)older version arena=text · category=exclude_ties · style_control=true | 1385.6 1379.9–1391.3 | community | LMArena Leaderboard Dataset | |
| Creative Writing (style control)older version arena=text · category=creative_writing · style_control=true | 1381.4 1372.1–1390.7 | community | LMArena Leaderboard Dataset | |
| Industry Entertainment And Sports And Media (style control)older version arena=text · category=industry_entertainment_and_sports_and_media · style_control=true | 1380.9 1372.8–1388.9 | community | LMArena Leaderboard Dataset | |
| Korean (style control)older version arena=text · category=korean · style_control=true | 1372.0 1348.6–1395.3 | community | LMArena Leaderboard Dataset |
knowledge science
| GPQA Diamond implementation=vals-ai | 81.6 % | independent | Vals AI | |
| GPQA Diamond implementation=vals-ai | thinking | 76.5 % | independent | Vals AI |
| MMLU-Pro implementation=vals-ai | 83.7 % | independent | Vals AI | |
| MMLU-Pro implementation=vals-ai | thinking | 83.7 % | independent | Vals AI |
multimodal
| MMMU implementation=vals-ai | thinking | 80.8 % | independent | Vals AI |
| MMMU implementation=vals-ai | 79.5 % | independent | Vals AI |
professional
| CorpFin | thinking | 59.8 % | independent | Vals AI |
| CorpFin | 59.0 % | independent | Vals AI | |
| LegalBench | thinking | 82.6 % | independent | Vals AI |
| LegalBench | 82.5 % | independent | Vals AI | |
| MedQA | 91.4 % | independent | Vals AI | |
| MedQA | thinking | 91.2 % | independent | Vals AI |
| TaxEval | 72.7 % | independent | Vals AI | |
| TaxEval | thinking | 72.4 % | independent | Vals AI |
reasoning math
| AIME implementation=vals-ai | thinking | 51.5 % | independent | Vals AI |
| AIME implementation=vals-ai | 49.8 % | independent | Vals AI |
