Model comparison
DeepSeek LLM 67B vs Qwen3 32B
Qwen3 32B is the stronger model overall, scoring 39.2 to 24.9 on the Noometry Index.
Last verified . 14 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 14 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and Qwen3 32B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3 32B leads 40.0 to 7.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.8% for DeepSeek LLM 67B and 66.9% for Qwen3 32B.
Side by side
| DeepSeek LLM 67B | Qwen3 32B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 24.9 | 39.2 |
| Released | 2023-11-29 | 2025-04 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.70 |
| Output $ / M tokens | — | $2.80 |
| Results tracked | 15 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 32B leads
DeepSeek LLM 67B: 31.9 (#278), Qwen3 32B: 37.7 (#190)
| Benchmark | DeepSeek LLM 67B | Qwen3 32B |
|---|---|---|
| LMArena Coding | 1096 | 1358 |
| Aider Polyglot | — | 40% |
| SciCode | — | 35.4% |
Agentic & Tool Use Not comparable
DeepSeek LLM 67B: —, Qwen3 32B: 32.6 (#62)
| Benchmark | DeepSeek LLM 67B | Qwen3 32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 48.7% |
Reasoning Qwen3 32B leads
DeepSeek LLM 67B: 16.5 (#304), Qwen3 32B: 20.2 (#241)
| Benchmark | DeepSeek LLM 67B | Qwen3 32B |
|---|---|---|
| Chess Puzzles | 0% | 5% |
| LMArena Hard Prompts | 1070 | 1334 |
| Epoch Capabilities Index | 110.5 | 138.51 |
| Kagi LLM Benchmark | — | 54.9% |
| CritPt | — | 0.3% |
| DTBench | — | 67.5% |
| LMCA | — | 17.3% |
Math Qwen3 32B leads
DeepSeek LLM 67B: 8.7 (#324), Qwen3 32B: 39.7 (#99)
| Benchmark | DeepSeek LLM 67B | Qwen3 32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 66.9% |
| LMArena Math | 1108 | 1399 |
| MATH Level 5 | 6.4% | — |
Knowledge Qwen3 32B leads
DeepSeek LLM 67B: 7.0 (#313), Qwen3 32B: 40.0 (#125)
| Benchmark | DeepSeek LLM 67B | Qwen3 32B |
|---|---|---|
| GPQA Diamond | 24.6% | 65.7% |
| Vectara Hallucination Rate | — | 5.9% |
| LMArena Expert | — | 1362 |
Multilingual Qwen3 32B leads
DeepSeek LLM 67B: 29.4 (#267), Qwen3 32B: 45.6 (#167)
| Benchmark | DeepSeek LLM 67B | Qwen3 32B |
|---|---|---|
| LMArena Non-English | 1073 | 1317 |
| LMArena Chinese | 1132 | 1357 |
| LMArena German | — | 1341 |
| LMArena Russian | — | 1311 |
Instruction Following Qwen3 32B leads
DeepSeek LLM 67B: 55.4 (#277), Qwen3 32B: 68.9 (#179)
| Benchmark | DeepSeek LLM 67B | Qwen3 32B |
|---|---|---|
| LMArena Instruction Following | 1079 | 1305 |
Long Context Qwen3 32B leads
DeepSeek LLM 67B: 33.1 (#265), Qwen3 32B: 43.8 (#87)
| Benchmark | DeepSeek LLM 67B | Qwen3 32B |
|---|---|---|
| LMArena Longer Query | 1092 | 1327 |
| Fiction.LiveBench | — | 74.2% |
Writing & Preference Qwen3 32B leads
DeepSeek LLM 67B: 31.6 (#282), Qwen3 32B: 52.9 (#163)
| Benchmark | DeepSeek LLM 67B | Qwen3 32B |
|---|---|---|
| LMArena Text | 1105 | 1340 |
| LMArena Creative Writing | 1067 | 1297 |
| LMArena Multi-Turn | 1082 | 1331 |
Frequently asked questions
Is DeepSeek LLM 67B better than Qwen3 32B?
Qwen3 32B is the stronger model overall, scoring 39.2 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Qwen3 32B better for coding?
Qwen3 32B scores higher on coding benchmarks: 37.7 versus 31.9 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Qwen3 32B share?
14 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Qwen3 32B has 26.