AI Agent Hub

Compare LLM Models

Select 2-5 models to compare side by side.

Select models to compare 0 / 5 selected
Top downloads
Select 2-5 models
Attribute
🤖 Qwen3-14B
Qwen
Company Qwen
Release Date 2025-04-28
Parameters (B) 14.8
Architecture Transformer
Context Window 128000
Input Modalities text
Open Source Yes
License apache-2.0
Score 0
VRAM (GB) 28
Compute Dense 14.8B-parameter model (~28 GB FP16 weights); recommended 28 GB+ VRAM for full-precision inference, or 8–16 GB with Q4/Q8 quantization on RTX 4070/4090-class GPUs
Benchmark: AIME-2024 79.3
Benchmark: AIME-2025 70.4
Benchmark: BBH 81.1
Benchmark: C-Eval 86.2
Benchmark: GPQA 39.9
Benchmark: GPQA-Diamond 64
Benchmark: GSM8K 92.5
Benchmark: IFEval 85.4
Benchmark: LiveCodeBench 63.5
Benchmark: MATH 62.0
Benchmark: MATH-500 96.8
Benchmark: MBPP 73.4
Benchmark: MMLU 81.0
Benchmark: MMLU-Pro 61.0
Pricing (Input) 0.1
Pricing (Output) 0.2