Model comparison
Qwen-14B vs StarCoder 2 7B
Qwen-14B has enough public results to be ranked (#275); StarCoder 2 7B does not yet, so treat this comparison as directional.
Last verified . 4 shared benchmarks.
Alibaba (Qwen)
NVIDIASummary
- They share 4 benchmarks with published results for both. Qwen-14B scores higher in 0 categories and StarCoder 2 7B in 1 category; one gap is clear of the uncertainty.
- The widest gap is in coding, where StarCoder 2 7B leads 34.8 to 31.2.
Side by side
| Qwen-14B | StarCoder 2 7B | |
|---|---|---|
| Provider | Alibaba (Qwen) | NVIDIA |
| Noometry Index | 31.4 | 37.0 |
| Released | 2023-09-24 | 2024-02-20 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 18 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding StarCoder 2 7B leads
Qwen-14B: 31.2 (#288), StarCoder 2 7B: 34.8
| Benchmark | Qwen-14B | StarCoder 2 7B |
|---|---|---|
| LMArena Coding | 1071 | — |
| BigCodeBench Complete | — | 27.7% |
| HumanEval+ | — | 29.9% |
Reasoning Not comparable
Qwen-14B: 19.6 (#257), StarCoder 2 7B: —
| Benchmark | Qwen-14B | StarCoder 2 7B |
|---|---|---|
| Epoch Capabilities Index | 113.03 | 93.56 |
| LMArena Hard Prompts | 1027 | — |
| BIG-Bench Hard | 55% | — |
| LAMBADA | 71.1% | — |
| PIQA | 79.9% | — |
| WinoGrande | — | 57.1% |
Math Not comparable
Qwen-14B: 31.2 (#227), StarCoder 2 7B: —
| Benchmark | Qwen-14B | StarCoder 2 7B |
|---|---|---|
| GSM8K | 61.3% | 32.7% |
| LMArena Math | 1068 | — |
Knowledge Not comparable
Qwen-14B: —, StarCoder 2 7B: —
| Benchmark | Qwen-14B | StarCoder 2 7B |
|---|---|---|
| ARC (AI2) Challenge | 84.4% | 38.7% |
| MMLU | 66.3% | 38.8% |
| BoolQ | 86.2% | — |
Multilingual Not comparable
Qwen-14B: 27.5 (#275), StarCoder 2 7B: —
| Benchmark | Qwen-14B | StarCoder 2 7B |
|---|---|---|
| LMArena Non-English | 1041 | — |
| LMArena Chinese | 1077 | — |
Instruction Following Not comparable
Qwen-14B: 52.4 (#289), StarCoder 2 7B: —
| Benchmark | Qwen-14B | StarCoder 2 7B |
|---|---|---|
| LMArena Instruction Following | 1031 | — |
Long Context Not comparable
Qwen-14B: 31.3 (#280), StarCoder 2 7B: —
| Benchmark | Qwen-14B | StarCoder 2 7B |
|---|---|---|
| LMArena Longer Query | 1028 | — |
Writing & Preference Not comparable
Qwen-14B: 27.6 (#299), StarCoder 2 7B: —
| Benchmark | Qwen-14B | StarCoder 2 7B |
|---|---|---|
| LMArena Text | 1051 | — |
| LMArena Creative Writing | 1028 | — |
| LMArena Multi-Turn | 1022 | — |
Frequently asked questions
Is Qwen-14B better than StarCoder 2 7B?
Qwen-14B has enough public results to be ranked (#275); StarCoder 2 7B does not yet, so treat this comparison as directional.
Is Qwen-14B or StarCoder 2 7B better for coding?
StarCoder 2 7B scores higher on coding benchmarks: 34.8 versus 31.2 in the Noometry coding category.
How many benchmarks do Qwen-14B and StarCoder 2 7B share?
4 benchmarks have published results for both models. Qwen-14B has 18 scored results on Noometry and StarCoder 2 7B has 7.