Compare · ModelsLive · 2 picked · head to head
Qwen2-72B vs DeepSeek V3
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
DeepSeek V3 wins on 4/4 benchmarks
DeepSeek V3 wins 4 of 4 shared benchmarks. Leads in knowledge · math · coding.
Category leads
knowledge·DeepSeek V3math·DeepSeek V3coding·DeepSeek V3
Hype vs Reality
Attention vs performance
Qwen2-72B
#177 by perf·no signal
DeepSeek V3
#59 by perf·no signal
Vendor risk
Mixed exposure
One or more vendors flagged
Alibaba (Qwen)
$293.0B·Tier 1
DeepSeek
$3.4B·Tier 1
Head to head
4 benchmarks · 2 models
Qwen2-72BDeepSeek V3
GPQA diamond
DeepSeek V3 leads by +21.0
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Qwen2-72B
21.0
DeepSeek V3
42.0
MATH level 5
DeepSeek V3 leads by +25.8
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
Qwen2-72B
39.1
DeepSeek V3
64.8
MMLU
DeepSeek V3 leads by +6.4
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
Qwen2-72B
76.5
DeepSeek V3
82.9
WeirdML
DeepSeek V3 leads by +24.8
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Qwen2-72B
11.3
DeepSeek V3
36.1
Full benchmark table
| Benchmark | Qwen2-72B | DeepSeek V3 |
|---|---|---|
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 21.0 | 42.0 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 39.1 | 64.8 |
MMLU Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge. | 76.5 | 82.9 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 11.3 | 36.1 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| — | — | — | — | |
| $0.20 | $0.80 | 131K tokens (~66 books) | $3.50 |