Model comparison
Qwen1.5-72B vs StarCoder 2 15B
Qwen1.5-72B has enough public results to be ranked (#285); 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. Qwen1.5-72B 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 31.9.
Side by side
| Qwen1.5-72B | StarCoder 2 15B | |
|---|---|---|
| Provider | Alibaba (Qwen) | NVIDIA |
| Noometry Index | 30.8 | 38.1 |
| Released | 2024-02-04 | 2024-02-20 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 22 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding StarCoder 2 15B leads
Qwen1.5-72B: 31.9 (#277), StarCoder 2 15B: 36.6
| Benchmark | Qwen1.5-72B | StarCoder 2 15B |
|---|---|---|
| BigCodeBench Complete | 40.3% | 38.4% |
| HumanEval+ | 59.1% | 37.8% |
| BigCodeBench Instruct | 33.2% | — |
| LMArena Coding | 1165 | — |
| MBPP+ | 61.6% | — |
Reasoning Not comparable
Qwen1.5-72B: 22.2 (#203), StarCoder 2 15B: —
| Benchmark | Qwen1.5-72B | StarCoder 2 15B |
|---|---|---|
| LMArena Hard Prompts | 1148 | — |
| Epoch Capabilities Index | — | 105 |
| WinoGrande | — | 64.3% |
Math Not comparable
Qwen1.5-72B: 33.2 (#205), StarCoder 2 15B: —
| Benchmark | Qwen1.5-72B | StarCoder 2 15B |
|---|---|---|
| LMArena Math | 1164 | — |
| GSM8K | — | 57.7% |
Knowledge Not comparable
Qwen1.5-72B: 11.5 (#300), StarCoder 2 15B: —
| Benchmark | Qwen1.5-72B | StarCoder 2 15B |
|---|---|---|
| GPQA Diamond | 28.8% | — |
| LMArena Expert | 1136 | — |
| ARC (AI2) Challenge | — | 47.2% |
| MMLU | — | 64.1% |
Multilingual Not comparable
Qwen1.5-72B: 33.2 (#253), StarCoder 2 15B: —
| Benchmark | Qwen1.5-72B | StarCoder 2 15B |
|---|---|---|
| LMArena Non-English | 1135 | — |
| LMArena Chinese | 1186 | — |
| LMArena French | 1159 | — |
| LMArena German | 1084 | — |
| LMArena Japanese | 1061 | — |
| LMArena Korean | 1050 | — |
| LMArena Russian | 1104 | — |
| LMArena Spanish | 1110 | — |
Instruction Following Not comparable
Qwen1.5-72B: 59.3 (#256), StarCoder 2 15B: —
| Benchmark | Qwen1.5-72B | StarCoder 2 15B |
|---|---|---|
| LMArena Instruction Following | 1141 | — |
Long Context Not comparable
Qwen1.5-72B: 35.1 (#243), StarCoder 2 15B: —
| Benchmark | Qwen1.5-72B | StarCoder 2 15B |
|---|---|---|
| LMArena Longer Query | 1157 | — |
Writing & Preference Not comparable
Qwen1.5-72B: 37.3 (#258), StarCoder 2 15B: —
| Benchmark | Qwen1.5-72B | StarCoder 2 15B |
|---|---|---|
| LMArena Text | 1166 | — |
| LMArena Creative Writing | 1137 | — |
| LMArena Multi-Turn | 1160 | — |
Frequently asked questions
Is Qwen1.5-72B better than StarCoder 2 15B?
Qwen1.5-72B has enough public results to be ranked (#285); StarCoder 2 15B does not yet, so treat this comparison as directional.
Is Qwen1.5-72B or StarCoder 2 15B better for coding?
StarCoder 2 15B scores higher on coding benchmarks: 36.6 versus 31.9 in the Noometry coding category.
How many benchmarks do Qwen1.5-72B and StarCoder 2 15B share?
2 benchmarks have published results for both models. Qwen1.5-72B has 22 scored results on Noometry and StarCoder 2 15B has 7.