Model comparison
DeepSeek LLM 67B vs Qwen3.8 27B
Qwen3.8 27B is the stronger model overall, scoring 45.3 to 24.8 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 0 categories and Qwen3.8 27B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3.8 27B leads 65.9 to 31.8.
Side by side
| DeepSeek LLM 67B | Qwen3.8 27B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 24.8 | 45.3 |
| Released | 2023-11-29 | 2026-08-14 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 33K |
| Input $ / M tokens | — | $0.99 |
| Output $ / M tokens | — | $1.49 |
| Results tracked | 15 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
DeepSeek LLM 67B: 31.3 (#273), Qwen3.8 27B: 50.3 (#42)
| Benchmark | DeepSeek LLM 67B | Qwen3.8 27B |
|---|---|---|
| LMArena Coding | 1096 | 1487 |
| LMArena WebDev | — | 1592 |
| SciCode | — | 46.6% |
Reasoning Qwen3.8 27B leads
DeepSeek LLM 67B: 15.9 (#300), Qwen3.8 27B: 31.5 (#71)
| Benchmark | DeepSeek LLM 67B | Qwen3.8 27B |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1459 |
| Epoch Capabilities Index | 110.54 | 149.38 |
| NYT Connections (extended) | — | 54.5% |
| CritPt | — | 5.4% |
| Chess Puzzles | 0% | — |
| DTBench | — | 88% |
| LMCA | — | 41.4% |
| Surface Evolver Bench | — | 45% |
Math Qwen3.8 27B leads
DeepSeek LLM 67B: 8.1 (#318), Qwen3.8 27B: 35.7 (#139)
| Benchmark | DeepSeek LLM 67B | Qwen3.8 27B |
|---|---|---|
| LMArena Math | 1108 | 1453 |
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| ProofBench | — | 16% |
| MATH Level 5 | 6.4% | — |
Knowledge Qwen3.8 27B leads
DeepSeek LLM 67B: 7.1 (#308), Qwen3.8 27B: 40.9 (#105)
| Benchmark | DeepSeek LLM 67B | Qwen3.8 27B |
|---|---|---|
| GPQA Diamond | 24.6% | — |
| LMArena Expert | — | 1482 |
Multimodal Not comparable
DeepSeek LLM 67B: —, Qwen3.8 27B: 40.5 (#36)
| Benchmark | DeepSeek LLM 67B | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | — | 1270 |
Multilingual Qwen3.8 27B leads
DeepSeek LLM 67B: 29.5 (#262), Qwen3.8 27B: 54.0 (#54)
| Benchmark | DeepSeek LLM 67B | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | 1073 | 1431 |
| LMArena Chinese | 1132 | 1504 |
| LMArena French | — | 1468 |
| LMArena German | — | 1435 |
| LMArena Japanese | — | 1392 |
| LMArena Korean | — | 1389 |
| LMArena Russian | — | 1421 |
| LMArena Spanish | — | 1444 |
Instruction Following Qwen3.8 27B leads
DeepSeek LLM 67B: 55.7 (#272), Qwen3.8 27B: 75.8 (#54)
| Benchmark | DeepSeek LLM 67B | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | 1079 | 1436 |
Long Context Qwen3.8 27B leads
DeepSeek LLM 67B: 33.1 (#260), Qwen3.8 27B: 44.3 (#66)
| Benchmark | DeepSeek LLM 67B | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | 1092 | 1448 |
Writing & Preference Qwen3.8 27B leads
DeepSeek LLM 67B: 31.8 (#277), Qwen3.8 27B: 65.9 (#41)
| Benchmark | DeepSeek LLM 67B | Qwen3.8 27B |
|---|---|---|
| LMArena Text | 1105 | 1441 |
| LMArena Creative Writing | 1067 | 1382 |
| LMArena Multi-Turn | 1082 | 1443 |
| EQ-Bench Creative Writing | — | 1671 |
Frequently asked questions
Is DeepSeek LLM 67B better than Qwen3.8 27B?
Qwen3.8 27B is the stronger model overall, scoring 45.3 to 24.8 on the Noometry Index.
Is DeepSeek LLM 67B or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.3 versus 31.3 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Qwen3.8 27B share?
11 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Qwen3.8 27B has 28.