Model comparison
Qwen2.5 72B Instruct vs StarCoder 2 15B
Qwen2.5 72B Instruct has enough public results to be ranked (#267); StarCoder 2 15B does not yet, so treat this comparison as directional.
Last verified . 5 shared benchmarks.
Alibaba (Qwen)
NVIDIASummary
- They share 5 benchmarks with published results for both. Qwen2.5 72B Instruct 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 33.2.
- The biggest single-benchmark swing is BigCodeBench Complete: 55.9% for Qwen2.5 72B Instruct and 38.4% for StarCoder 2 15B.
Side by side
| Qwen2.5 72B Instruct | StarCoder 2 15B | |
|---|---|---|
| Provider | Alibaba (Qwen) | NVIDIA |
| Noometry Index | 31.9 | 38.1 |
| Released | 2024-09 | 2024-02-20 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 8K | — |
| Input $ / M tokens | $1.40 | — |
| Output $ / M tokens | $5.60 | — |
| Results tracked | 43 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding StarCoder 2 15B leads
Qwen2.5 72B Instruct: 33.2 (#260), StarCoder 2 15B: 36.6
| Benchmark | Qwen2.5 72B Instruct | StarCoder 2 15B |
|---|---|---|
| BigCodeBench Complete | 55.9% | 38.4% |
| WeirdML | 16% | — |
| BigCodeBench Instruct | 45.8% | — |
| LMArena Coding | 1292 | — |
| HumanEval+ | — | 37.8% |
Agentic & Tool Use Not comparable
Qwen2.5 72B Instruct: 22.1 (#133), StarCoder 2 15B: —
| Benchmark | Qwen2.5 72B Instruct | StarCoder 2 15B |
|---|---|---|
| TheAgentCompany | 5.7% | — |
| BALROG | 16.2% | — |
| METR Time Horizons | 35.8% | — |
Reasoning Not comparable
Qwen2.5 72B Instruct: 22.3 (#199), StarCoder 2 15B: —
| Benchmark | Qwen2.5 72B Instruct | StarCoder 2 15B |
|---|---|---|
| Epoch Capabilities Index | 129 | 105 |
| WinoGrande | 82.3% | 64.3% |
| LMArena Hard Prompts | 1271 | — |
| DTBench | 62.9% | — |
| LMCA | 13.4% | — |
| BIG-Bench Hard | 79.8% | — |
| ForecastBench | 57.5 | — |
| HellaSwag | 84.8% | — |
| PIQA | 82.6% | — |
Math Not comparable
Qwen2.5 72B Instruct: 19.3 (#287), StarCoder 2 15B: —
| Benchmark | Qwen2.5 72B Instruct | StarCoder 2 15B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.1% | — |
| Omni-MATH | 33% | — |
| LMArena Math | 1283 | — |
| MATH Level 5 | 63.2% | — |
| GSM8K | — | 57.7% |
Knowledge Not comparable
Qwen2.5 72B Instruct: 27.0 (#253), StarCoder 2 15B: —
| Benchmark | Qwen2.5 72B Instruct | StarCoder 2 15B |
|---|---|---|
| ARC (AI2) Challenge | 94.5% | 47.2% |
| MMLU | 85.3% | 64.1% |
| GPQA Diamond | 49.1% | — |
| MMLU-Pro | 63.1% | — |
| Confabulations | 19.1% | — |
| GPQA (HELM) | 42.6% | — |
| LMArena Expert | 1245 | — |
| TriviaQA | 71.9% | — |
Multilingual Not comparable
Qwen2.5 72B Instruct: 41.0 (#213), StarCoder 2 15B: —
| Benchmark | Qwen2.5 72B Instruct | StarCoder 2 15B |
|---|---|---|
| LMArena Non-English | 1252 | — |
| LMArena Chinese | 1272 | — |
| LMArena French | 1280 | — |
| LMArena German | 1234 | — |
| LMArena Japanese | 1180 | — |
| LMArena Korean | 1188 | — |
| LMArena Russian | 1264 | — |
| LMArena Spanish | 1256 | — |
Instruction Following Not comparable
Qwen2.5 72B Instruct: 65.5 (#221), StarCoder 2 15B: —
| Benchmark | Qwen2.5 72B Instruct | StarCoder 2 15B |
|---|---|---|
| IFEval | 80.6% | — |
| LMArena Instruction Following | 1254 | — |
Long Context Not comparable
Qwen2.5 72B Instruct: 38.9 (#188), StarCoder 2 15B: —
| Benchmark | Qwen2.5 72B Instruct | StarCoder 2 15B |
|---|---|---|
| LMArena Longer Query | 1282 | — |
Writing & Preference Not comparable
Qwen2.5 72B Instruct: 46.7 (#215), StarCoder 2 15B: —
| Benchmark | Qwen2.5 72B Instruct | StarCoder 2 15B |
|---|---|---|
| LMArena Text | 1269 | — |
| LMArena Creative Writing | 1221 | — |
| WildBench | 80.2% | — |
| LMArena Multi-Turn | 1272 | — |
Frequently asked questions
Is Qwen2.5 72B Instruct better than StarCoder 2 15B?
Qwen2.5 72B Instruct has enough public results to be ranked (#267); StarCoder 2 15B does not yet, so treat this comparison as directional.
Is Qwen2.5 72B Instruct or StarCoder 2 15B better for coding?
StarCoder 2 15B scores higher on coding benchmarks: 36.6 versus 33.2 in the Noometry coding category.
How many benchmarks do Qwen2.5 72B Instruct and StarCoder 2 15B share?
5 benchmarks have published results for both models. Qwen2.5 72B Instruct has 43 scored results on Noometry and StarCoder 2 15B has 7.