HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/Qwen/Qwen/Qwen2.5-1.5B

Qwen/Qwen2.5-1.5B

family qwen · 10 variants · unknown

provider
Qwen
released
—
huggingface
https://huggingface.co/Qwen/Qwen2.5-1.5B
best scores
GPQA: 31.6MMLU-Pro: 47.5Humanity's Last Exam: 11.8LiveCodeBench: 28.6SciCode: 100.0AIME 2024: 55.5AIME 2025: 45.6

Best scores per benchmark

benchmarkcategorybest score
SciCodecoding100.0
AIME 2024math55.5
MMLU-Proknowledge47.5
AIME 2025math45.6
GPQAreasoning31.6
LiveCodeBenchcoding28.6
Humanity's Last Examknowledge11.8

Variants (10)

variantkindquantbitsvRAMdlorgyear
Qwen2.5-1.5B-Instructoriginalbf1616.013,344,230Qwen2024
Qwen2.5-1.5Boriginalbf1616.0913,187Qwen2024
Qwen2.5-Coder-1.5Boriginalbf1616.0266,156Qwen2024
Qwen2.5-Math-1.5Boriginalbf1616.0144,412Qwen2024
Qwen2.5-1.5B-unsloth-bnb-4bitbnb-4bit4bit60,677unsloth2025
Qwen2.5-1.5Boriginalbf1616.07,712unsloth2025
Qwen2.5-1.5B-bnb-4bitbnb-4bit4bit5,761unsloth2025
OpenReasoning-Nemotron-1.5Boriginalbf1616.02,738nvidia2025
Qwen2.5-1.5B-GGUFggufgguf3.01,292mradermacher2025
Qwen2.5-1.5B-Instruct-GGUFggufgguf16.0411Mungert2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.