Rankings / DeepSeek
Deepseek V3
released 2024-12-26
BenchAtlas Index
as of 2026-07-14
base configuration
rank #378
6 families · 3 categories · medium
Benchmark evidence
39 results
coding
| LiveCodeBench v5older version problems=558 · window_start=2024-08-01 | 34.6 % | independent | LiveCodeBench Leaderboard | |
| LiveCodeBench v6 problems=454 · window_start=2024-08-01 | 27.2 % | independent | LiveCodeBench Leaderboard |
external indices
| ECI | 133.0 points | independent | Epoch AI Benchmarking Hub |
human preference
| Industry Medicine And Healthcare (style control)older version arena=text · category=industry_medicine_and_healthcare · style_control=true | 1392.9 1374.7–1411.0 | community | LMArena Leaderboard Dataset | |
| Coding (style control)older version arena=text · category=coding · style_control=true | 1387.7 1377.6–1397.8 | community | LMArena Leaderboard Dataset | |
| Industry Software And It Services (style control)older version arena=text · category=industry_software_and_it_services · style_control=true | 1384.4 1376.0–1392.8 | community | LMArena Leaderboard Dataset | |
| Industry Legal And Government (style control)older version arena=text · category=industry_legal_and_government · style_control=true | 1384.0 1367.6–1400.4 | community | LMArena Leaderboard Dataset | |
| Longer Query (style control)older version arena=text · category=longer_query · style_control=true | 1374.7 1364.1–1385.2 | community | LMArena Leaderboard Dataset | |
| Multi Turn (style control)older version arena=text · category=multi_turn · style_control=true | 1373.7 1363.9–1383.5 | 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 | 1372.7 1363.1–1382.4 | community | LMArena Leaderboard Dataset | |
| English (style control)older version arena=text · category=english · style_control=true | 1370.4 1364.5–1376.4 | community | LMArena Leaderboard Dataset | |
| Chinese (style control)older version arena=text · category=chinese · style_control=true | 1369.1 1353.8–1384.4 | community | LMArena Leaderboard Dataset | |
| Hard Prompts English (style control)older version arena=text · category=hard_prompts_english · style_control=true | 1362.2 1352.5–1371.9 | community | LMArena Leaderboard Dataset | |
| Overall (style control) arena=text · category=overall · style_control=true | 1358.5 1353.8–1363.2 | community | LMArena Leaderboard Dataset | |
| Russian (style control)older version arena=text · category=russian · style_control=true | 1356.2 1344.7–1367.6 | 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 | 1356.0 1347.9–1364.1 | 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 | 1350.3 1337.9–1362.6 | community | LMArena Leaderboard Dataset | |
| Hard Prompts (style control)older version arena=text · category=hard_prompts · style_control=true | 1350.3 1342.2–1358.4 | community | LMArena Leaderboard Dataset | |
| Creative Writing (style control)older version arena=text · category=creative_writing · style_control=true | 1348.9 1338.8–1358.9 | 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 | 1348.8 1339.5–1358.1 | community | LMArena Leaderboard Dataset | |
| Expert (style control)older version arena=text · category=expert · style_control=true | 1347.0 1330.3–1363.6 | community | LMArena Leaderboard Dataset | |
| Instruction Following (style control)older version arena=text · category=instruction_following · style_control=true | 1343.8 1336.9–1350.6 | community | LMArena Leaderboard Dataset | |
| Non English (style control)older version arena=text · category=non_english · style_control=true | 1342.6 1335.9–1349.2 | community | LMArena Leaderboard Dataset | |
| Exclude Ties (style control)older version arena=text · category=exclude_ties · style_control=true | 1327.8 1320.4–1335.2 | community | LMArena Leaderboard Dataset | |
| Industry Mathematical (style control)older version arena=text · category=industry_mathematical · style_control=true | 1317.5 1305.7–1329.2 | community | LMArena Leaderboard Dataset | |
| Math (style control)older version arena=text · category=math · style_control=true | 1310.9 1300.1–1321.6 | community | LMArena Leaderboard Dataset | |
| Japanese (style control)older version arena=text · category=japanese · style_control=true | 1288.1 1264.3–1311.8 | community | LMArena Leaderboard Dataset |
knowledge science
| GPQA Diamond | 56.5 % | independent | Epoch AI Benchmarking Hub 2025-01-27 | |
| GPQA Diamond implementation=vals-ai | 54.5 % | independent | Vals AI | |
| MMLU-Pro implementation=vals-ai | 73.8 % | independent | Vals AI |
professional
| CorpFin | 52.5 % | independent | Vals AI | |
| LegalBench | 80.8 % | independent | Vals AI | |
| MedQA | 80.9 % | independent | Vals AI | |
| TaxEval | 67.9 % | independent | Vals AI |
reasoning math
| AIME implementation=vals-ai | 27.5 % | independent | Vals AI | |
| FrontierMath | 1.7 % | independent | Epoch AI Benchmarking Hub 2025-03-07 | |
| MATH-500 implementation=vals-ai | 80.4 % | independent | Vals AI | |
| MATH Level 5 | 64.8 % | independent | Epoch AI Benchmarking Hub 2025-01-27 | |
| OTIS Mock AIME 2024–2025 | 15.8 % | independent | Epoch AI Benchmarking Hub 2025-02-25 |
Agent + model results
systems, not bare-model scores
| agent + model KGCompass + Deepseek V3 | SWE-bench Lite | 36.7 % | unverified | SWE-bench Leaderboard |
These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.