Model comparison
DeepSeek LLM 67B vs Qwen1.5-32B
Qwen1.5-32B is the stronger model overall, scoring 30.5 to 24.9 on the Noometry Index.
Last verified . 11 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 11 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 1 category and Qwen1.5-32B in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen1.5-32B leads 33.0 to 8.7.
- The biggest single-benchmark swing is GPQA Diamond: 24.6% for DeepSeek LLM 67B and 30.7% for Qwen1.5-32B.
Side by side
| DeepSeek LLM 67B | Qwen1.5-32B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 24.9 | 30.5 |
| Released | 2023-11-29 | 2024-02-04 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 15 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek LLM 67B: 31.9 (#278), Qwen1.5-32B: 31.7 (#282)
| Benchmark | DeepSeek LLM 67B | Qwen1.5-32B |
|---|---|---|
| LMArena Coding | 1096 | 1155 |
| BigCodeBench Instruct | — | 32.3% |
| BigCodeBench Complete | — | 42% |
Reasoning Qwen1.5-32B leads
DeepSeek LLM 67B: 16.5 (#304), Qwen1.5-32B: 21.8 (#212)
| Benchmark | DeepSeek LLM 67B | Qwen1.5-32B |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1130 |
| Chess Puzzles | 0% | — |
| Epoch Capabilities Index | 110.5 | — |
Math Qwen1.5-32B leads
DeepSeek LLM 67B: 8.7 (#324), Qwen1.5-32B: 33.0 (#207)
| Benchmark | DeepSeek LLM 67B | Qwen1.5-32B |
|---|---|---|
| LMArena Math | 1108 | 1155 |
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| MATH Level 5 | 6.4% | — |
Knowledge Qwen1.5-32B leads
DeepSeek LLM 67B: 7.0 (#313), Qwen1.5-32B: 13.5 (#296)
| Benchmark | DeepSeek LLM 67B | Qwen1.5-32B |
|---|---|---|
| GPQA Diamond | 24.6% | 30.7% |
| LMArena Expert | — | 1126 |
| MMLU | — | 74.4% |
Multilingual Qwen1.5-32B leads
DeepSeek LLM 67B: 29.4 (#267), Qwen1.5-32B: 31.4 (#259)
| Benchmark | DeepSeek LLM 67B | Qwen1.5-32B |
|---|---|---|
| LMArena Non-English | 1073 | 1106 |
| LMArena Chinese | 1132 | 1177 |
| LMArena French | — | 1101 |
| LMArena German | — | 1058 |
| LMArena Japanese | — | 1027 |
| LMArena Korean | — | 1008 |
| LMArena Russian | — | 1073 |
| LMArena Spanish | — | 1089 |
Instruction Following Qwen1.5-32B leads
DeepSeek LLM 67B: 55.4 (#277), Qwen1.5-32B: 57.7 (#265)
| Benchmark | DeepSeek LLM 67B | Qwen1.5-32B |
|---|---|---|
| LMArena Instruction Following | 1079 | 1116 |
Long Context Qwen1.5-32B leads
DeepSeek LLM 67B: 33.1 (#265), Qwen1.5-32B: 34.7 (#246)
| Benchmark | DeepSeek LLM 67B | Qwen1.5-32B |
|---|---|---|
| LMArena Longer Query | 1092 | 1146 |
Writing & Preference Qwen1.5-32B leads
DeepSeek LLM 67B: 31.6 (#282), Qwen1.5-32B: 34.2 (#271)
| Benchmark | DeepSeek LLM 67B | Qwen1.5-32B |
|---|---|---|
| LMArena Text | 1105 | 1137 |
| LMArena Creative Writing | 1067 | 1083 |
| LMArena Multi-Turn | 1082 | 1140 |
Frequently asked questions
Is DeepSeek LLM 67B better than Qwen1.5-32B?
Qwen1.5-32B is the stronger model overall, scoring 30.5 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Qwen1.5-32B better for coding?
They score almost the same on coding (31.9 vs 31.7); test both on your own repository before choosing.
How many benchmarks do DeepSeek LLM 67B and Qwen1.5-32B share?
11 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Qwen1.5-32B has 21.