Model comparison
Qwen1.5-32B vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 30.5 on the Noometry Index.
Last verified . 15 shared benchmarks.
Alibaba (Qwen)Summary
- They share 15 benchmarks with published results for both. Qwen1.5-32B scores higher in 2 categories and Qwen2.5-Coder-32B in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5-Coder-32B leads 33.4 to 13.5.
- The biggest single-benchmark swing is BigCodeBench Instruct: 32.3% for Qwen1.5-32B and 49% for Qwen2.5-Coder-32B.
Side by side
| Qwen1.5-32B | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 30.5 | 33.4 |
| Released | 2024-02-04 | 2024-09-18 |
| Weights | Open | Open |
| Context window | — | 33K |
| Max output | — | 29K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1 |
| Results tracked | 21 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen1.5-32B leads
Qwen1.5-32B: 31.7 (#282), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Qwen1.5-32B | Qwen2.5-Coder-32B |
|---|---|---|
| BigCodeBench Instruct | 32.3% | 49% |
| LMArena Coding | 1155 | 1276 |
| BigCodeBench Complete | 42% | 58% |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| LiveBench Coding | — | 56.9% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Reasoning Too close to call
Qwen1.5-32B: 21.8 (#212), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Qwen1.5-32B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1130 | 1251 |
| LiveBench Reasoning | — | 42.1% |
| LiveBench Data Analysis | — | 49.9% |
| Epoch Capabilities Index | — | 119.49 |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Too close to call
Qwen1.5-32B: 33.0 (#207), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Qwen1.5-32B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1155 | 1251 |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge Qwen2.5-Coder-32B leads
Qwen1.5-32B: 13.5 (#296), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Qwen1.5-32B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1126 | 1221 |
| MMLU | 74.4% | 79.1% |
| GPQA Diamond | 30.7% | — |
| ARC (AI2) Challenge | — | 70.5% |
Multilingual Qwen2.5-Coder-32B leads
Qwen1.5-32B: 31.4 (#259), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Qwen1.5-32B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1106 | 1205 |
| LMArena Chinese | 1177 | 1222 |
| LMArena Russian | 1073 | 1228 |
| LMArena French | 1101 | — |
| LMArena German | 1058 | — |
| LMArena Japanese | 1027 | — |
| LMArena Korean | 1008 | — |
| LMArena Spanish | 1089 | — |
Instruction Following Qwen2.5-Coder-32B leads
Qwen1.5-32B: 57.7 (#265), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Qwen1.5-32B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1116 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Qwen2.5-Coder-32B leads
Qwen1.5-32B: 34.7 (#246), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Qwen1.5-32B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1146 | 1251 |
Writing & Preference Qwen2.5-Coder-32B leads
Qwen1.5-32B: 34.2 (#271), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Qwen1.5-32B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1137 | 1230 |
| LMArena Creative Writing | 1083 | 1174 |
| LMArena Multi-Turn | 1140 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Qwen1.5-32B better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 30.5 on the Noometry Index.
Is Qwen1.5-32B or Qwen2.5-Coder-32B better for coding?
Qwen1.5-32B scores higher on coding benchmarks: 31.7 versus 22.6 in the Noometry coding category.
How many benchmarks do Qwen1.5-32B and Qwen2.5-Coder-32B share?
15 benchmarks have published results for both models. Qwen1.5-32B has 21 scored results on Noometry and Qwen2.5-Coder-32B has 31.