What are the hardware requirements for Qwen 14B?

Qwen 14B requires approximately 28.4 GB of VRAM and supports a context window of 8K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen 14B and how large is it?

Qwen 14B is developed by Qwen, a model with 14b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Qwen 14B perform on standard benchmarks?

Qwen 14B has the following published scores: MMLU 67.7. Compare against reference models on this page or on the LLM Explorer leaderboards.

Qwen 14B by Qwen — VRAM 28.4GB, 8K context

Name: Qwen 14B
Author: Qwen

Qwen 14B is an open-source language model by Qwen. Features: 14b LLM, VRAM: 28.4GB, Context: 8K, LLM Explorer Score: 0.12, Arc: 58.3, HellaSwag: 84, MMLU: 67.7, GSM8K: 59.

Arxiv:2309.16609 Custom code En Qwen Region:us Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen-14B

Qwen 14B Parameters and Internals

Model Type

text generation, AI assistant

Use Cases

Areas:

research, commercial applications

Additional Notes

Vocabulary consists of over 150K tokens, supports a variety of languages and improves performance on multilingual tasks.

Supported Languages

Chinese (high quality), English (high quality), multilingual (high quality), code (high quality), mathematics (high quality)

Training Details

Data Sources:

web texts, books, codes, encyclopedias, mathematics

Data Volume:

over 3 trillion tokens

Context Length:

2048

Model Architecture:

Transformer with RoPE relative position encoding, SwiGLU activation function, and RMSNorm

Input Output

Accepted Modalities:

text

Performance Tips:

Optional installation of flash-attention for acceleration.

LLM Name	Qwen 14B
Repository 🤗	https://huggingface.co/Qwen/Qwen-14B
Model Size	14b
Required VRAM	28.4 GB
Updated	2026-07-31
Maintainer	Qwen
Model Type	qwen
Model Files	2.0 GB: 1-of-15 2.0 GB: 2-of-15 1.9 GB: 3-of-15 1.9 GB: 4-of-15 1.9 GB: 5-of-15 1.9 GB: 6-of-15 1.9 GB: 7-of-15 1.9 GB: 8-of-15 1.9 GB: 9-of-15 1.9 GB: 10-of-15 1.9 GB: 11-of-15 1.9 GB: 12-of-15 1.9 GB: 13-of-15 1.9 GB: 14-of-15 1.6 GB: 15-of-15
Supported Languages	zh en
Model Architecture	QWenLMHeadModel
Context Length	8192
Model Max Length	8192
Transformers Version	4.32.0
Tokenizer Class	QWenTokenizer
Vocabulary Size	152064

Rank the Qwen 14B capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Qwen 14B

Best Alternatives	Context / RAM	Downloads	Likes
Qwen 14B Chat	8K / 28.4 GB	3837	373
Qwen 14B Chat Int4	8K / 9.6 GB	642	100
Qwen 14B Chat Int8	8K / 15.9 GB	118	7

Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen-14B.