GSM8K#10 of 28
exact_match
View run →84.9%±1.0%
Evaluated across 13 benchmarks. Ranks in the top 3 on 0 of 13. Strongest showing on GSM8K (84.9% exact_match, #10 of 28). Weakest on MBPP(Instruct) (0.0% pass@1, #8 of 24).
| Benchmark | Metric | Score | Rank | Actions |
|---|---|---|---|---|
| GSM8K | exact_match | 84.9%±1.0% | #10 of 28 | View run → |
| IFEval | prompt_level_strict_acc | 81.7%±1.7% | #11 of 28 | View run → |
| EQ-Bench | eqbench | 75.8±1.9 | #17 of 28 | View run → |
| MMLU | acc | 73.0%±0.4% | #12 of 28 | View run → |
| MBPP | pass@1 | 65.6%±2.1% | #8 of 28 | View run → |
| MMLU-Pro | exact_match | 57.7%±0.4% | #17 of 28 | View run → |
| MGSM | exact_match | 47.9%±0.8% | #26 of 28 | View run → |
| LongBench | aggregate | 32.1%±0.4% | #4 of 17 | View run → |
| GPQA Diamond | acc | 27.8%±3.2% | #20 of 28 | View run → |
| GPQA Main | acc | 25.4%±2.1% | #23 of 28 | View run → |
| GPQA Extended | acc | 23.8%±1.8% | #26 of 28 | View run → |
| BBH | exact_match | 13.2%±0.3% | #17 of 28 | View run → |
| MBPP(Instruct) | pass@1 | 0.0%±0.0% | #8 of 24 | View run → |
Citation
FrozeBench. "Qwen/Qwen3-8B." https://frozebench.com/models/qwen-qwen3-8b. Retrieved 2026-06-04.
BibTeX
@misc{frozebench_Qwen_Qwen3_8B,
title = {Qwen/Qwen3-8B},
howpublished = {\url{https://frozebench.com/models/qwen-qwen3-8b}},
year = {2026},
note = {FrozeBench. Retrieved 2026-06-04.}
}URL
https://frozebench.com/models/qwen-qwen3-8b