Model comparison
Qwen-14B vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 31.4 on the Noometry Index.
Last verified . 14 shared benchmarks.
Alibaba (Qwen)Summary
- They share 14 benchmarks with published results for both. Qwen-14B scores higher in 1 category and Qwen2.5-Coder-32B in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2.5-Coder-32B leads 41.6 to 27.6.
Side by side
| Qwen-14B | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 31.4 | 33.4 |
| Released | 2023-09-24 | 2024-09-18 |
| Weights | Open | Open |
| Context window | — | 33K |
| Max output | — | 29K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1 |
| Results tracked | 18 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen-14B leads
Qwen-14B: 31.2 (#288), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Qwen-14B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1071 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Reasoning Qwen2.5-Coder-32B leads
Qwen-14B: 19.6 (#257), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Qwen-14B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1027 | 1251 |
| Epoch Capabilities Index | 113.03 | 119.49 |
| LiveBench Reasoning | — | 42.1% |
| LiveBench Data Analysis | — | 49.9% |
| BIG-Bench Hard | 55% | — |
| HellaSwag | — | 83% |
| LAMBADA | 71.1% | — |
| LiveBench | — | 46.2% |
| PIQA | 79.9% | — |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
Qwen-14B: 31.2 (#227), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Qwen-14B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1068 | 1251 |
| GSM8K | 61.3% | 93% |
| LiveBench Math | — | 46.6% |
Knowledge Not comparable
Qwen-14B: —, Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Qwen-14B | Qwen2.5-Coder-32B |
|---|---|---|
| ARC (AI2) Challenge | 84.4% | 70.5% |
| MMLU | 66.3% | 79.1% |
| LMArena Expert | — | 1221 |
| BoolQ | 86.2% | — |
Multilingual Qwen2.5-Coder-32B leads
Qwen-14B: 27.5 (#275), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Qwen-14B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1041 | 1205 |
| LMArena Chinese | 1077 | 1222 |
| LMArena Russian | — | 1228 |
Instruction Following Qwen2.5-Coder-32B leads
Qwen-14B: 52.4 (#289), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Qwen-14B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1031 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Qwen2.5-Coder-32B leads
Qwen-14B: 31.3 (#280), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Qwen-14B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1028 | 1251 |
Writing & Preference Qwen2.5-Coder-32B leads
Qwen-14B: 27.6 (#299), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Qwen-14B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1051 | 1230 |
| LMArena Creative Writing | 1028 | 1174 |
| LMArena Multi-Turn | 1022 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Qwen-14B better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 31.4 on the Noometry Index.
Is Qwen-14B or Qwen2.5-Coder-32B better for coding?
Qwen-14B scores higher on coding benchmarks: 31.2 versus 22.6 in the Noometry coding category.
How many benchmarks do Qwen-14B and Qwen2.5-Coder-32B share?
14 benchmarks have published results for both models. Qwen-14B has 18 scored results on Noometry and Qwen2.5-Coder-32B has 31.