Rankings / DeepSeek
Deepseek V3 0324
released 2025-03-24
BenchAtlas Index
as of 2026-09-11
base configuration
rank #363
5 families · 3 categories · medium
base configuration
rank #377
12 families · 5 categories · high
Benchmark evidence
56 results
agentic coding
| τ²-Bench | 47.1 % | independent | Artificial Analysis | |
| τ²-Bench subset=banking | 4.7 % | independent | Artificial Analysis |
coding
| LiveCodeBench v6 implementation=artificial-analysis | 40.5 % | independent | Artificial Analysis | |
| SciCode | 39.0 % | independent | Artificial Analysis |
external indices
| AA Coding Index | 21.2 points | independent | Artificial Analysis | |
| Intelligence Index v4.1 | 9.7 points | independent | Artificial Analysis | |
| AA Math Index | 41.0 points | independent | Artificial Analysis | |
| ECI | 136.0 points | independent | Epoch AI Benchmarking Hub |
human preference
| Coding (style control)older version arena=text · category=coding · style_control=true | 1428.6 1421.8–1435.4 | community | LMArena Leaderboard Dataset | |
| Industry Medicine And Healthcare (style control)older version arena=text · category=industry_medicine_and_healthcare · style_control=true | 1426.6 1414.8–1438.3 | community | LMArena Leaderboard Dataset | |
| Industry Software And It Services (style control)older version arena=text · category=industry_software_and_it_services · style_control=true | 1423.5 1417.9–1429.2 | community | LMArena Leaderboard Dataset | |
| Hard Prompts English (style control)older version arena=text · category=hard_prompts_english · style_control=true | 1420.4 1413.7–1427.1 | community | LMArena Leaderboard Dataset | |
| French (style control)older version arena=text · category=french · style_control=true | 1416.2 1391.4–1441.0 | community | LMArena Leaderboard Dataset | |
| Chinese (style control)older version arena=text · category=chinese · style_control=true | 1415.7 1404.7–1426.7 | 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 | 1414.8 1407.6–1422.0 | community | LMArena Leaderboard Dataset | |
| Multi Turn (style control)older version arena=text · category=multi_turn · style_control=true | 1411.2 1404.0–1418.4 | community | LMArena Leaderboard Dataset | |
| Industry Legal And Government (style control)older version arena=text · category=industry_legal_and_government · style_control=true | 1411.1 1399.9–1422.3 | community | LMArena Leaderboard Dataset | |
| English (style control)older version arena=text · category=english · style_control=true | 1409.0 1404.0–1414.0 | community | LMArena Leaderboard Dataset | |
| Hard Prompts (style control)older version arena=text · category=hard_prompts · style_control=true | 1408.1 1402.9–1413.4 | community | LMArena Leaderboard Dataset | |
| Russian (style control)older version arena=text · category=russian · style_control=true | 1399.7 1390.1–1409.2 | community | LMArena Leaderboard Dataset | |
| Expert (style control)older version arena=text · category=expert · style_control=true | 1398.9 1386.3–1411.4 | community | LMArena Leaderboard Dataset | |
| Polish (style control)older version arena=text · category=polish · style_control=true | 1396.1 1384.0–1408.2 | community | LMArena Leaderboard Dataset | |
| Overall (style control) arena=text · category=overall · style_control=true | 1395.7 1391.8–1399.6 | community | LMArena Leaderboard Dataset | |
| Longer Query (style control)older version arena=text · category=longer_query · style_control=true | 1393.7 1387.0–1400.5 | community | LMArena Leaderboard Dataset | |
| German (style control)older version arena=text · category=german · style_control=true | 1391.3 1373.6–1409.0 | 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 | 1389.7 1382.1–1397.2 | community | LMArena Leaderboard Dataset | |
| Creative Writing (style control)older version arena=text · category=creative_writing · style_control=true | 1389.5 1381.4–1397.6 | 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 | 1386.2 1379.7–1392.6 | 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 | 1382.2 1375.2–1389.2 | community | LMArena Leaderboard Dataset | |
| Non English (style control)older version arena=text · category=non_english · style_control=true | 1379.1 1374.3–1383.9 | community | LMArena Leaderboard Dataset | |
| Instruction Following (style control)older version arena=text · category=instruction_following · style_control=true | 1378.2 1372.3–1384.1 | community | LMArena Leaderboard Dataset | |
| Spanish (style control)older version arena=text · category=spanish · style_control=true | 1378.0 1358.4–1397.6 | community | LMArena Leaderboard Dataset | |
| Exclude Ties (style control)older version arena=text · category=exclude_ties · style_control=true | 1376.7 1371.2–1382.1 | community | LMArena Leaderboard Dataset | |
| Math (style control)older version arena=text · category=math · style_control=true | 1368.5 1358.1–1378.8 | community | LMArena Leaderboard Dataset | |
| Industry Mathematical (style control)older version arena=text · category=industry_mathematical · style_control=true | 1366.2 1355.4–1377.1 | community | LMArena Leaderboard Dataset | |
| Japanese (style control)older version arena=text · category=japanese · style_control=true | 1344.1 1325.0–1363.2 | community | LMArena Leaderboard Dataset | |
| Korean (style control)older version arena=text · category=korean · style_control=true | 1339.5 1319.7–1359.3 | community | LMArena Leaderboard Dataset |
knowledge science
| GPQA Diamond | 67.6 % | independent | Epoch AI Benchmarking Hub 2025-04-01 | |
| GPQA Diamond implementation=artificial-analysis | 65.5 % | independent | Artificial Analysis | |
| GPQA Diamond implementation=vals-ai | 61.6 % | independent | Vals AI | |
| Humanity's Last Exam implementation=artificial-analysis | 4.7 % | independent | Artificial Analysis | |
| MMLU-Pro implementation=artificial-analysis | 81.9 % | independent | Artificial Analysis | |
| MMLU-Pro implementation=vals-ai | 79.5 % | independent | Vals AI |
long context instruction
| AA-LCR | 40.7 % | independent | Artificial Analysis | |
| IFBench | 41.0 % | independent | Artificial Analysis |
professional
| CorpFin | 54.7 % | independent | Vals AI | |
| LegalBench | 77.7 % | independent | Vals AI | |
| MedQA | 82.0 % | independent | Vals AI | |
| TaxEval | 71.1 % | independent | Vals AI |
reasoning math
| AIME implementation=vals-ai | 52.2 % | independent | Vals AI | |
| AIME implementation=artificial-analysis | 52.0 % | independent | Artificial Analysis | |
| AIME year=2025 · implementation=artificial-analysis | 41.0 % | independent | Artificial Analysis | |
| MATH-500 implementation=artificial-analysis | 94.2 % | independent | Artificial Analysis | |
| MATH-500 implementation=vals-ai | 88.6 % | independent | Vals AI | |
| MATH Level 5 | 75.5 % | independent | Epoch AI Benchmarking Hub 2025-04-01 | |
| OTIS Mock AIME 2024–2025 | 37.8 % | independent | Epoch AI Benchmarking Hub 2025-04-01 |
Agent + model results
systems, not bare-model scores
| agent + model SWE-Exp + Deepseek V3 0324 | SWE-bench Verified | 42.0 % | unverified | SWE-bench Leaderboard |
| agent + model Artificial Analysis harness + Deepseek V3 0324 | Terminal-Bench Hard | 15.2 % | independent | Artificial Analysis |
| agent + model Artificial Analysis harness + Deepseek V3 0324 | Terminal-Bench 2.1 | 13.9 % | independent | Artificial Analysis |
These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.