Model comparison
Qwen-14B vs Qwen1.5-32B
Qwen-14B and Qwen1.5-32B score almost the same on the Noometry Index (30.8 vs 30.0), so choose on price, context window or the category you care about most.
Last verified . 11 shared benchmarks.
Alibaba (Qwen)Summary
- They share 11 benchmarks with published results for both. Qwen-14B scores higher in 0 categories and Qwen1.5-32B in 7 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen1.5-32B leads 34.4 to 27.8.
Side by side
| Qwen-14B | Qwen1.5-32B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 30.8 | 30.0 |
| Released | 2023-09-24 | 2024-02-04 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 18 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Qwen-14B: 30.6 (#283), Qwen1.5-32B: 31.1 (#278)
| Benchmark | Qwen-14B | Qwen1.5-32B |
|---|---|---|
| LMArena Coding | 1072 | 1155 |
| BigCodeBench Instruct | — | 32.3% |
| BigCodeBench Complete | — | 42% |
Reasoning Qwen1.5-32B leads
Qwen-14B: 18.7 (#255), Qwen1.5-32B: 20.9 (#210)
| Benchmark | Qwen-14B | Qwen1.5-32B |
|---|---|---|
| LMArena Hard Prompts | 1027 | 1131 |
| BIG-Bench Hard | 55% | — |
| Epoch Capabilities Index | 112.8 | — |
| LAMBADA | 71.1% | — |
| PIQA | 79.9% | — |
Math Qwen1.5-32B leads
Qwen-14B: 28.7 (#224), Qwen1.5-32B: 30.5 (#207)
| Benchmark | Qwen-14B | Qwen1.5-32B |
|---|---|---|
| LMArena Math | 1068 | 1155 |
| GSM8K | 61.3% | — |
Knowledge Not comparable
Qwen-14B: —, Qwen1.5-32B: 13.5 (#291)
| Benchmark | Qwen-14B | Qwen1.5-32B |
|---|---|---|
| MMLU | 66.3% | 74.4% |
| GPQA Diamond | — | 30.7% |
| LMArena Expert | — | 1126 |
| ARC (AI2) Challenge | 84.4% | — |
| BoolQ | 86.2% | — |
Multilingual Qwen1.5-32B leads
Qwen-14B: 27.7 (#270), Qwen1.5-32B: 31.6 (#254)
| Benchmark | Qwen-14B | Qwen1.5-32B |
|---|---|---|
| LMArena Non-English | 1042 | 1106 |
| LMArena Chinese | 1077 | 1176 |
| LMArena French | — | 1100 |
| LMArena German | — | 1058 |
| LMArena Japanese | — | 1027 |
| LMArena Korean | — | 1007 |
| LMArena Russian | — | 1073 |
| LMArena Spanish | — | 1089 |
Instruction Following Qwen1.5-32B leads
Qwen-14B: 52.6 (#284), Qwen1.5-32B: 58.0 (#260)
| Benchmark | Qwen-14B | Qwen1.5-32B |
|---|---|---|
| LMArena Instruction Following | 1031 | 1116 |
Long Context Qwen1.5-32B leads
Qwen-14B: 31.3 (#276), Qwen1.5-32B: 34.7 (#241)
| Benchmark | Qwen-14B | Qwen1.5-32B |
|---|---|---|
| LMArena Longer Query | 1029 | 1146 |
Writing & Preference Qwen1.5-32B leads
Qwen-14B: 27.8 (#294), Qwen1.5-32B: 34.4 (#266)
| Benchmark | Qwen-14B | Qwen1.5-32B |
|---|---|---|
| LMArena Text | 1051 | 1137 |
| LMArena Creative Writing | 1028 | 1083 |
| LMArena Multi-Turn | 1022 | 1140 |
Frequently asked questions
Is Qwen-14B better than Qwen1.5-32B?
Qwen-14B and Qwen1.5-32B score almost the same on the Noometry Index (30.8 vs 30.0), so choose on price, context window or the category you care about most.
Is Qwen-14B or Qwen1.5-32B better for coding?
They score almost the same on coding (30.6 vs 31.1); test both on your own repository before choosing.
How many benchmarks do Qwen-14B and Qwen1.5-32B share?
11 benchmarks have published results for both models. Qwen-14B has 18 scored results on Noometry and Qwen1.5-32B has 21.