GSM8K#7 of 28
exact_match
View run →87.6%±0.9%
Evaluated across 13 benchmarks. Ranks in the top 3 on 0 of 13. Strongest showing on GSM8K (87.6% exact_match, #7 of 28). Weakest on MBPP(Instruct) (0.0% pass@1, #23 of 24).
| Benchmark | Metric | Score | Rank | Actions |
|---|---|---|---|---|
| GSM8K | exact_match | 87.6%±0.9% | #7 of 28 | View run → |
| IFEval | prompt_level_strict_acc | 80.2%±1.7% | #13 of 28 | View run → |
| EQ-Bench | eqbench | 79.3±1.8 | #9 of 28 | View run → |
| MMLU | acc | 77.2%±0.3% | #9 of 28 | View run → |
| MBPP | pass@1 | 71.4%±2.0% | #5 of 28 | View run → |
| MGSM | exact_match | 66.2%±0.8% | #13 of 28 | View run → |
| MMLU-Pro | exact_match | 65.3%±0.4% | #11 of 28 | View run → |
| BBH | exact_match | 41.8%±0.5% | #11 of 28 | View run → |
| GPQA Diamond | acc | 29.3%±3.2% | #14 of 28 | View run → |
| GPQA Extended | acc | 26.2%±1.9% | #20 of 28 | View run → |
| GPQA Main | acc | 26.1%±2.1% | #19 of 28 | View run → |
| LongBench | aggregate | 8.1%±0.1% | #16 of 17 | View run → |
| MBPP(Instruct) | pass@1 | 0.0%±0.0% | #23 of 24 | View run → |
Citation
FrozeBench. "Qwen/Qwen3-14B." https://frozebench.com/models/qwen-qwen3-14b. Retrieved 2026-06-04.
BibTeX
@misc{frozebench_Qwen_Qwen3_14B,
title = {Qwen/Qwen3-14B},
howpublished = {\url{https://frozebench.com/models/qwen-qwen3-14b}},
year = {2026},
note = {FrozeBench. Retrieved 2026-06-04.}
}URL
https://frozebench.com/models/qwen-qwen3-14b