HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/Qwen/Qwen/Qwen3-14B

Qwen/Qwen3-14B

family qwen · 22 variants · unknown

provider
Qwen
released
—
huggingface
https://huggingface.co/Qwen/Qwen3-14B
best scores
GPQA: 64.0MMLU: 88.6AIME 2024: 79.3DeepSWE: 42.2

Best scores per benchmark

benchmarkcategorybest score
MMLUknowledge88.6
AIME 2024math79.3
GPQAreasoning64.0
DeepSWEswe42.2

Variants (22)

variantkindquantbitsvRAMdlorgyear
Qwen3-14B-AWQawqawq2,391,865Qwen2025
Qwen3-14B-FP8fp8fp88.0461,461Qwen2025
Qwen3-14B-unsloth-bnb-4bitbnb-4bit4bit220,094unsloth2025
Qwen3-14B-NVFP4nvfp4nvfp44.0123,890nvidia2025
Qwen3-14B-GGUFggufgguf4.091,154Qwen2025
Qwen3-14B-FP8fp8fp88.074,381nvidia2025
Qwen3-14B-GGUFggufgguf3.045,818lmstudio-community2025
Qwen3-14B-MLX-4bitmlxmlx45,562lmstudio-community2025
Qwen3-14B-MLX-8bitmlxmlx40,767lmstudio-community2025
Qwen3-14B-GGUFggufgguf16.033,216unsloth2025
Qwen_Qwen3-14B-GGUFggufgguf2.031,176bartowski2025
Qwen3-14B-bnb-4bitbnb-4bit4bit9,125unsloth2025
Qwen3-14Boriginalbf1616.08,531unsloth2025
Qwen3-14B-GGUFggufgguf4.02,973ggml-org2025
Qwen3-14B-128K-GGUFggufgguf16.02,776unsloth2025
Qwen3-14B-abliteratedabliteratedabliterated2,626mlabonne2025
Qwen3-14B-abliterated-GGUFggufgguf16.0879Mungert2025
MagenticBrainoriginalbf1616.0276microsoft2026
FrogMini-14B-2510originalbf1616.0181microsoft2026
Hermes-4-14B-GGUFggufgguf16.0155Mungert2025
DMind-1-mini-GGUFggufgguf16.087Mungert2025
Qwen3-Nemotron-14B-BRRMoriginalbf1616.051nvidia2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.