HF Model & Benchmark Matrix

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

Qwen/Qwen2.5-14B-Instruct

family qwen · 16 variants · unknown

provider
Qwen
released
—
huggingface
https://huggingface.co/Qwen/Qwen2.5-14B-Instruct
best scores
LiveCodeBench: 65.6GPQA: 31.6MMLU-Pro: 47.5Humanity's Last Exam: 11.8SciCode: 2.2AIME 2024: 55.5AIME 2025: 45.6

Best scores per benchmark

benchmarkcategorybest score
LiveCodeBenchcoding65.6
AIME 2024math55.5
MMLU-Proknowledge47.5
AIME 2025math45.6
GPQAreasoning31.6
Humanity's Last Examknowledge11.8
SciCodecoding2.2

Variants (16)

variantkindquantbitsvRAMdlorgyear
Qwen2.5-14B-Instruct-AWQawqawq2,375,721Qwen2024
Qwen2.5-14B-Instructoriginalbf1616.0425,048unsloth2025
Qwen2.5-14B-Instruct-GGUFggufgguf2.094,273bartowski2024
Qwen2.5-14B-Instruct-GPTQ-Int4gptqgptq43,329Qwen2024
Qwen2.5-14B-Instruct-GGUFggufgguf31,359Qwen2024
Qwen2.5-14B-Instruct-unsloth-bnb-4bitbnb-4bit4bit28,614unsloth2025
Qwen2.5-14B-Instruct-bnb-4bitbnb-4bit4bit28,527unsloth2025
Qwen2.5-14B-Instruct-GGUFggufgguf3.010,063lmstudio-community2024
Qwen2.5-14B-Instruct-GGUFggufgguf3.04,844mradermacher2025
Qwen2.5-14B-Instruct-MLX-4bitmlxmlx2,725lmstudio-community2024
Qwen2.5-14B-Instruct-GPTQ-Int8gptqgptq1,830Qwen2024
Qwen2.5-14B-Instruct-MLX-8bitmlxmlx1,606lmstudio-community2024
OpenCodeReasoning-Nemotron-14B-GGUFggufgguf16.0187Mungert2025
OpenCodeReasoning-Nemotron-14Boriginalbf1616.0128nvidia2025
OpenReasoning-Nemotron-14B-GGUFggufgguf16.078Mungert2025
OpenCodeReasoning-Nemotron-1.1-14Boriginalbf1616.073nvidia2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.