Model comparison
DeepSeek LLM 67B vs Qwen3 8B
Qwen3 8B is the stronger model overall, scoring 33.7 to 24.9 on the Noometry Index.
Last verified . 4 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 4 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and Qwen3 8B in 5 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3 8B leads 36.1 to 7.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.8% for DeepSeek LLM 67B and 56.1% for Qwen3 8B.
Side by side
| DeepSeek LLM 67B | Qwen3 8B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 24.9 | 33.7 |
| Released | 2023-11-29 | 2025-04 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $0.18 |
| Output $ / M tokens | — | $0.70 |
| Results tracked | 15 | 11 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 8B leads
DeepSeek LLM 67B: 31.9 (#278), Qwen3 8B: 34.0 (#248)
| Benchmark | DeepSeek LLM 67B | Qwen3 8B |
|---|---|---|
| SciCode | — | 22.6% |
| LMArena Coding | 1096 | — |
Agentic & Tool Use Not comparable
DeepSeek LLM 67B: —, Qwen3 8B: 30.2 (#78)
| Benchmark | DeepSeek LLM 67B | Qwen3 8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 42.6% |
Reasoning Too close to call
DeepSeek LLM 67B: 16.5 (#304), Qwen3 8B: 16.6 (#303)
| Benchmark | DeepSeek LLM 67B | Qwen3 8B |
|---|---|---|
| Chess Puzzles | 0% | 5% |
| Epoch Capabilities Index | 110.5 | 136.17 |
| CritPt | — | 0% |
| LMArena Hard Prompts | 1070 | — |
| DTBench | — | 59.7% |
| LMCA | — | 8.8% |
Math Qwen3 8B leads
DeepSeek LLM 67B: 8.7 (#324), Qwen3 8B: 34.9 (#191)
| Benchmark | DeepSeek LLM 67B | Qwen3 8B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 56.1% |
| LMArena Math | 1108 | — |
| MATH Level 5 | 6.4% | — |
Knowledge Qwen3 8B leads
DeepSeek LLM 67B: 7.0 (#313), Qwen3 8B: 36.1 (#173)
| Benchmark | DeepSeek LLM 67B | Qwen3 8B |
|---|---|---|
| GPQA Diamond | 24.6% | 56.8% |
| Vectara Hallucination Rate | — | 4.8% |
Multilingual Not comparable
DeepSeek LLM 67B: 29.4 (#267), Qwen3 8B: —
| Benchmark | DeepSeek LLM 67B | Qwen3 8B |
|---|---|---|
| LMArena Non-English | 1073 | — |
| LMArena Chinese | 1132 | — |
Instruction Following Not comparable
DeepSeek LLM 67B: 55.4 (#277), Qwen3 8B: —
| Benchmark | DeepSeek LLM 67B | Qwen3 8B |
|---|---|---|
| LMArena Instruction Following | 1079 | — |
Long Context Qwen3 8B leads
DeepSeek LLM 67B: 33.1 (#265), Qwen3 8B: 37.9 (#210)
| Benchmark | DeepSeek LLM 67B | Qwen3 8B |
|---|---|---|
| Fiction.LiveBench | — | 62.1% |
| LMArena Longer Query | 1092 | — |
Writing & Preference Not comparable
DeepSeek LLM 67B: 31.6 (#282), Qwen3 8B: —
| Benchmark | DeepSeek LLM 67B | Qwen3 8B |
|---|---|---|
| LMArena Text | 1105 | — |
| LMArena Creative Writing | 1067 | — |
| LMArena Multi-Turn | 1082 | — |
Frequently asked questions
Is DeepSeek LLM 67B better than Qwen3 8B?
Qwen3 8B is the stronger model overall, scoring 33.7 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Qwen3 8B better for coding?
Qwen3 8B scores higher on coding benchmarks: 34.0 versus 31.9 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Qwen3 8B share?
4 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Qwen3 8B has 11.