DeepSeek-V3 0324
Zero-eval
by DeepSeek
About
DeepSeek-V3 0324 is a language model developed by DeepSeek. It achieves strong performance with an average score of 70.4% across 5 benchmarks. It excels particularly in MATH-500 (94.0%), MMLU-Pro (81.2%), GPQA (68.4%). Released in 2025, it represents DeepSeek's latest advancement in AI technology.
Timeline
AnnouncedMar 25, 2025
ReleasedMar 25, 2025
Specifications
Training Tokens14.8T
License & Family
License
MIT + Model License (Commercial use allowed)
Benchmark Performance Overview
Performance metrics and category breakdown
Overall Performance
5 benchmarks
Average Score
70.4%
Best Score
94.0%
High Performers (80%+)
2Top Categories
math
94.0%
general
69.7%
code
49.2%
Benchmark Performance
Top benchmark scores with normalized values (0-100%)
Ranking Across Benchmarks
Position relative to other models on each benchmark
MATH-500
Rank #12 of 22
#9DeepSeek R1 Distill Qwen 32B
94.3%
#10DeepSeek R1 Distill Llama 70B
94.5%
#11Phi 4 Mini Reasoning
94.6%
#12DeepSeek-V3 0324
94.0%
#13DeepSeek R1 Distill Qwen 14B
93.9%
#14DeepSeek R1 Distill Qwen 7B
92.8%
#15QwQ-32B
90.6%
MMLU-Pro
Rank #4 of 60
#1Qwen3-235B-A22B-Instruct-2507
83.0%
#2DeepSeek-R1
84.0%
#3DeepSeek-R1-0528
85.0%
#4DeepSeek-V3 0324
81.2%
#5Kimi K2 Instruct
81.1%
#6Llama 4 Maverick
80.5%
#7Claude 3.5 Sonnet
77.6%
GPQA
Rank #33 of 115
#30Phi 4 Reasoning Plus
68.9%
#31GPT-4.5
69.5%
#32Llama 4 Maverick
69.8%
#33DeepSeek-V3 0324
68.4%
#34Magistral Small 2506
68.2%
#35Claude 3.5 Sonnet
67.2%
#36Llama-3.3 Nemotron Super 49B v1
66.7%
AIME 2024
Rank #32 of 41
#29Kimi K2 Instruct
69.6%
#30Magistral Small 2506
70.7%
#31Gemini 2.0 Flash Thinking
73.3%
#32DeepSeek-V3 0324
59.4%
#33DeepSeek R1 Distill Qwen 1.5B
52.7%
#34QwQ-32B-Preview
50.0%
#35GPT-4.1 mini
49.6%
LiveCodeBench
Rank #23 of 44
#20QwQ-32B-Preview
50.0%
#21DeepSeek R1 Zero
50.0%
#22Magistral Medium
50.3%
#23DeepSeek-V3 0324
49.2%
#24Llama 4 Maverick
43.4%
#25DeepSeek R1 Distill Llama 8B
39.6%
#26DeepSeek-V3
37.6%
All Benchmark Results for DeepSeek-V3 0324
Complete list of benchmark scores with detailed information
MATH-500 MATH-500 benchmark | math | text | 0.94 | 94.0% | Self-reported |
MMLU-Pro MMLU-Pro benchmark | general | text | 0.81 | 81.2% | Self-reported |
GPQA GPQA benchmark | general | text | 0.68 | 68.4% | Self-reported |
AIME 2024 AIME 2024 benchmark | general | text | 0.59 | 59.4% | Self-reported |
LiveCodeBench LiveCodeBench benchmark | code | text | 0.49 | 49.2% | Self-reported |