Qwen 3 14B
by Alibaba · qwen-3 family
14B
parameters
text-generation code-generation reasoning multilingual math tool-use creative-writing summarization
Qwen 3 14B is one of the strongest mid-range models available, excelling at coding, math, reasoning, and creative tasks. Hybrid thinking mode lets it match the performance of much larger models on complex problems while staying fast on simple ones. At Q4 it fits on 16 GB GPUs or Macs, making it a top pick for the RTX 4060 Ti 16GB, RTX 5070 Ti, or M-series Macs with 16 GB. A strong contender for the best daily driver at this VRAM tier.
Quick Start with Ollama
ollama run 14b-q4_K_M | Creator | Alibaba |
| Parameters | 14B |
| Architecture | transformer-decoder |
| Context | 128K tokens |
| Released | Apr 29, 2025 |
| License | Apache 2.0 |
| Ollama | qwen3:14b |
Quantization Options
| Format | File Size | VRAM Required | Quality | Ollama Tag |
|---|---|---|---|---|
| Q4_K_M rec | 9.3 GB | 12 GB | | 14b-q4_K_M |
| Q8_0 | 15.5 GB | 19 GB | | 14b-q8_0 |
| F16 | 29 GB | 33 GB | | 14b-fp16 |
Compatible Hardware
Q4_K_M requires 12 GB VRAM
Benchmark Scores
80.5
mmlu