Qwen·released 2024-09-181 source
Qwen2.5-Coder-14B benchmark scores: 5 benchmarks tracked. 0% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 5 benchmarks·best result 88.7 on GSM8K (self-reported)·0 of 5 independently reproduced
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| GSM8K | 88.7accuracy (%) | self-reported· optimizedT1 | 2024-09-18Qwen2.5-Coder Technical Report |
| HellaSwag | 73.6accuracy (%) | self-reported· optimizedT1 | 2024-09-18Qwen2.5-Coder Technical Report |
| MMLU | 66.93accuracy (%) | self-reported· optimizedT1 | 2024-09-18Qwen2.5-Coder Technical Report |
| ARC AI2 | 54.67accuracy (%) | self-reported· optimizedT1 | 2024-09-18Qwen2.5-Coder Technical Report |
| Winogrande | 53.6accuracy (%) | self-reported· optimizedT1 | 2024-09-18Qwen2.5-Coder Technical Report |
Claims drawn from cited facts, not live model generation.
Developed by Qwen and released in 2024, this model is a mid-size 14-billion-parameter model. This scale implies a balance between the efficiency and lower cost of small models and the greater headroom and expense of large ones.
4 cited facts
This model is tracked across five benchmarks and does not currently hold a top score on any of them. Because these results are self-reported and not independently confirmed, they should be read as claims rather than verified facts.
3 cited facts
Relative to the best reported scores, this model trails the leader by an average of 20.9 points, a wide gap that places it well behind the front-runners on its benchmarks. It holds the top score on none of the tracked benchmarks (0 leaders), so it has not achieved category leadership in this suite.
2 cited facts
2 facts cross-checked across data sources: 2 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: curated_param_counts, epoch_benchmarks
Qwen2.5-Coder-14B is an AI model developed by Qwen, released 2024-09-18. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
Qwen2.5-Coder-14B has recorded scores on 5 benchmarks, each shown with its evidence status.
Qwen2.5-Coder-14B has recorded scores on 5 benchmarks — Winogrande, MMLU, ARC AI2, GSM8K, HellaSwag. The full table above shows each score with its evidence status.
0 of 5 recorded scores (0%) are independently reproduced rather than self-reported by the lab.
Qwen2.5-Coder-14B has 5 tracked claims: 5 self-reported.