Model comparison
DeepSeek LLM 67B vs Qwen2-72B
Qwen2-72B is the stronger model overall, scoring 30.0 to 24.9 on the Noometry Index.
Last verified . 13 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 1 category and Qwen2-72B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2-72B leads 30.2 to 8.7.
- The biggest single-benchmark swing is MATH Level 5: 6.4% for DeepSeek LLM 67B and 39.1% for Qwen2-72B.
Side by side
| DeepSeek LLM 67B | Qwen2-72B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 24.9 | 30.0 |
| Released | 2023-11-29 | 2024-06-07 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 15 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek LLM 67B leads
DeepSeek LLM 67B: 31.9 (#278), Qwen2-72B: 29.1 (#310)
| Benchmark | DeepSeek LLM 67B | Qwen2-72B |
|---|---|---|
| LMArena Coding | 1096 | 1196 |
| WeirdML | — | 11.3% |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |
Agentic & Tool Use Not comparable
DeepSeek LLM 67B: —, Qwen2-72B: 17.0 (#146)
| Benchmark | DeepSeek LLM 67B | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Qwen2-72B leads
DeepSeek LLM 67B: 16.5 (#304), Qwen2-72B: 23.2 (#181)
| Benchmark | DeepSeek LLM 67B | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1191 |
| Epoch Capabilities Index | 110.5 | 125.28 |
| Chess Puzzles | 0% | — |
Math Qwen2-72B leads
DeepSeek LLM 67B: 8.7 (#324), Qwen2-72B: 30.2 (#236)
| Benchmark | DeepSeek LLM 67B | Qwen2-72B |
|---|---|---|
| LMArena Math | 1108 | 1235 |
| MATH Level 5 | 6.4% | 39.1% |
| OTIS Mock AIME 2024-2025 | 0.8% | — |
Knowledge Qwen2-72B leads
DeepSeek LLM 67B: 7.0 (#313), Qwen2-72B: 21.2 (#275)
| Benchmark | DeepSeek LLM 67B | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 24.6% | 40.8% |
| LMArena Expert | — | 1171 |
| MMLU | — | 82.4% |
Multilingual Qwen2-72B leads
DeepSeek LLM 67B: 29.4 (#267), Qwen2-72B: 35.9 (#244)
| Benchmark | DeepSeek LLM 67B | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1073 | 1176 |
| LMArena Chinese | 1132 | 1240 |
| LMArena French | — | 1170 |
| LMArena German | — | 1151 |
| LMArena Japanese | — | 1111 |
| LMArena Korean | — | 1083 |
| LMArena Russian | — | 1169 |
| LMArena Spanish | — | 1169 |
Instruction Following Qwen2-72B leads
DeepSeek LLM 67B: 55.4 (#277), Qwen2-72B: 61.7 (#241)
| Benchmark | DeepSeek LLM 67B | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1079 | 1181 |
Long Context Qwen2-72B leads
DeepSeek LLM 67B: 33.1 (#265), Qwen2-72B: 36.1 (#235)
| Benchmark | DeepSeek LLM 67B | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1092 | 1192 |
Writing & Preference Qwen2-72B leads
DeepSeek LLM 67B: 31.6 (#282), Qwen2-72B: 40.8 (#241)
| Benchmark | DeepSeek LLM 67B | Qwen2-72B |
|---|---|---|
| LMArena Text | 1105 | 1203 |
| LMArena Creative Writing | 1067 | 1181 |
| LMArena Multi-Turn | 1082 | 1196 |
Frequently asked questions
Is DeepSeek LLM 67B better than Qwen2-72B?
Qwen2-72B is the stronger model overall, scoring 30.0 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Qwen2-72B better for coding?
DeepSeek LLM 67B scores higher on coding benchmarks: 31.9 versus 29.1 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Qwen2-72B share?
13 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Qwen2-72B has 26.