Model comparison
DeepSeek-V3.1-Terminus vs Qwen1.5-14B
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 32.7 on the Noometry Index.
Last verified . 10 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 10 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 7 categories and Qwen1.5-14B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.0 to 33.6.
Side by side
| DeepSeek-V3.1-Terminus | Qwen1.5-14B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 43.1 | 32.7 |
| Released | 2025-09-22 | 2024-02-04 |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 147K | — |
| Input $ / M tokens | $0.27 | — |
| Output $ / M tokens | $1 | — |
| Results tracked | 16 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 42.0 (#113), Qwen1.5-14B: 33.1 (#263)
| Benchmark | DeepSeek-V3.1-Terminus | Qwen1.5-14B |
|---|---|---|
| LMArena Coding | 1426 | 1138 |
| SciCode | 40.6% | — |
| ALE-Bench | 745.17 | — |
Reasoning DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 26.4 (#133), Qwen1.5-14B: 21.4 (#223)
| Benchmark | DeepSeek-V3.1-Terminus | Qwen1.5-14B |
|---|---|---|
| LMArena Hard Prompts | 1426 | 1113 |
| Kagi LLM Benchmark | 57.4% | — |
| CritPt | 1.7% | — |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |
Math DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 38.5 (#137), Qwen1.5-14B: 32.4 (#215)
| Benchmark | DeepSeek-V3.1-Terminus | Qwen1.5-14B |
|---|---|---|
| LMArena Math | 1402 | 1125 |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Qwen1.5-14B: 29.8 (#232)
| Benchmark | DeepSeek-V3.1-Terminus | Qwen1.5-14B |
|---|---|---|
| LMArena Expert | — | 1094 |
| MMLU | — | 68.6% |
Multilingual DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 52.1 (#92), Qwen1.5-14B: 30.7 (#262)
| Benchmark | DeepSeek-V3.1-Terminus | Qwen1.5-14B |
|---|---|---|
| LMArena Non-English | 1407 | 1095 |
| LMArena Russian | 1436 | 1046 |
| LMArena Chinese | — | 1147 |
| LMArena French | — | 1116 |
| LMArena German | — | 1043 |
| LMArena Japanese | — | 1019 |
| LMArena Spanish | — | 1085 |
Instruction Following DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 74.0 (#106), Qwen1.5-14B: 56.8 (#271)
| Benchmark | DeepSeek-V3.1-Terminus | Qwen1.5-14B |
|---|---|---|
| LMArena Instruction Following | 1404 | 1102 |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 43.4 (#97), Qwen1.5-14B: 33.7 (#257)
| Benchmark | DeepSeek-V3.1-Terminus | Qwen1.5-14B |
|---|---|---|
| LMArena Longer Query | 1421 | 1113 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Qwen1.5-14B: 33.6 (#276)
| Benchmark | DeepSeek-V3.1-Terminus | Qwen1.5-14B |
|---|---|---|
| LMArena Text | 1419 | 1128 |
| LMArena Creative Writing | 1403 | 1091 |
| LMArena Multi-Turn | 1411 | 1110 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Qwen1.5-14B?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 32.7 on the Noometry Index.
Is DeepSeek-V3.1-Terminus or Qwen1.5-14B better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 42.0 versus 33.1 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1-Terminus and Qwen1.5-14B share?
10 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Qwen1.5-14B has 17.