Model comparison
Qwen1.5-32B vs StarCoder 2 15B
Qwen1.5-32B has enough public results to be ranked (#293); StarCoder 2 15B does not yet, so treat this comparison as directional.
Last verified . 2 shared benchmarks.
Alibaba (Qwen)
NVIDIASummary
- They share 2 benchmarks with published results for both. Qwen1.5-32B scores higher in 0 categories and StarCoder 2 15B in 1 category; one gap is clear of the uncertainty.
- The widest gap is in coding, where StarCoder 2 15B leads 36.6 to 31.7.
Side by side
| Qwen1.5-32B | StarCoder 2 15B | |
|---|---|---|
| Provider | Alibaba (Qwen) | NVIDIA |
| Noometry Index | 30.5 | 38.1 |
| Released | 2024-02-04 | 2024-02-20 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 21 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding StarCoder 2 15B leads
Qwen1.5-32B: 31.7 (#282), StarCoder 2 15B: 36.6
| Benchmark | Qwen1.5-32B | StarCoder 2 15B |
|---|---|---|
| BigCodeBench Complete | 42% | 38.4% |
| BigCodeBench Instruct | 32.3% | — |
| LMArena Coding | 1155 | — |
| HumanEval+ | — | 37.8% |
Reasoning Not comparable
Qwen1.5-32B: 21.8 (#212), StarCoder 2 15B: —
| Benchmark | Qwen1.5-32B | StarCoder 2 15B |
|---|---|---|
| LMArena Hard Prompts | 1130 | — |
| Epoch Capabilities Index | — | 105 |
| WinoGrande | — | 64.3% |
Math Not comparable
Qwen1.5-32B: 33.0 (#207), StarCoder 2 15B: —
| Benchmark | Qwen1.5-32B | StarCoder 2 15B |
|---|---|---|
| LMArena Math | 1155 | — |
| GSM8K | — | 57.7% |
Knowledge Not comparable
Qwen1.5-32B: 13.5 (#296), StarCoder 2 15B: —
| Benchmark | Qwen1.5-32B | StarCoder 2 15B |
|---|---|---|
| MMLU | 74.4% | 64.1% |
| GPQA Diamond | 30.7% | — |
| LMArena Expert | 1126 | — |
| ARC (AI2) Challenge | — | 47.2% |
Multilingual Not comparable
Qwen1.5-32B: 31.4 (#259), StarCoder 2 15B: —
| Benchmark | Qwen1.5-32B | StarCoder 2 15B |
|---|---|---|
| LMArena Non-English | 1106 | — |
| LMArena Chinese | 1177 | — |
| LMArena French | 1101 | — |
| LMArena German | 1058 | — |
| LMArena Japanese | 1027 | — |
| LMArena Korean | 1008 | — |
| LMArena Russian | 1073 | — |
| LMArena Spanish | 1089 | — |
Instruction Following Not comparable
Qwen1.5-32B: 57.7 (#265), StarCoder 2 15B: —
| Benchmark | Qwen1.5-32B | StarCoder 2 15B |
|---|---|---|
| LMArena Instruction Following | 1116 | — |
Long Context Not comparable
Qwen1.5-32B: 34.7 (#246), StarCoder 2 15B: —
| Benchmark | Qwen1.5-32B | StarCoder 2 15B |
|---|---|---|
| LMArena Longer Query | 1146 | — |
Writing & Preference Not comparable
Qwen1.5-32B: 34.2 (#271), StarCoder 2 15B: —
| Benchmark | Qwen1.5-32B | StarCoder 2 15B |
|---|---|---|
| LMArena Text | 1137 | — |
| LMArena Creative Writing | 1083 | — |
| LMArena Multi-Turn | 1140 | — |
Frequently asked questions
Is Qwen1.5-32B better than StarCoder 2 15B?
Qwen1.5-32B has enough public results to be ranked (#293); StarCoder 2 15B does not yet, so treat this comparison as directional.
Is Qwen1.5-32B or StarCoder 2 15B better for coding?
StarCoder 2 15B scores higher on coding benchmarks: 36.6 versus 31.7 in the Noometry coding category.
How many benchmarks do Qwen1.5-32B and StarCoder 2 15B share?
2 benchmarks have published results for both models. Qwen1.5-32B has 21 scored results on Noometry and StarCoder 2 15B has 7.