What are the hardware requirements for Qwen2.5 Coder 14B?

Qwen2.5 Coder 14B requires approximately 29.7 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen2.5 Coder 14B and how large is it?

Qwen2.5 Coder 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.

Where can I download or evaluate Qwen2.5 Coder 14B?

Qwen2.5 Coder 14B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen2.5 Coder 14B by Qwen — VRAM 29.7GB, 32K context

Name: Qwen2.5 Coder 14B
Rating: 2 (2 reviews)
Author: Qwen

Qwen2.5 Coder 14B is an open-source language model by Qwen. Features: 14b LLM, VRAM: 29.7GB, Context: 32K, License: apache-2.0, Code Generating, LLM Explorer Score: 0.29.

Arxiv:2309.00071 Arxiv:2407.10671 Arxiv:2409.12186 Base model:finetune:qwen/qwen2... Base model:qwen/qwen2.5-14b Code Codegen Codeqwen Conversational Deploy:azure En Endpoints compatible Qwen Qwen-coder Qwen2 Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen2.5-Coder-14B

Qwen2.5 Coder 14B Benchmarks

LLME Score: 0.28616

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

Qwen2.5 Coder 14B (Qwen/Qwen2.5-Coder-14B)

🌟 Advertise your project 🚀

Qwen2.5 Coder 14B Parameters and Internals

Model Type

text-generation

Use Cases

Applications:

code generation, code reasoning, code fixing

Limitations:

not recommended for conversational tasks

Additional Notes

The model has long-context support allowing up to 128K tokens processed.

Supported Languages

en (high)

Training Details

Data Sources:

source code, text-code grounding, Synthetic data

Data Volume:

5.5 trillion tokens

Methodology:

transformers with RoPE, SwiGLU, RMSNorm, and Attention QKV bias

Context Length:

131072

Model Architecture:

transformers with RoPE, SwiGLU, RMSNorm, and Attention QKV bias

Input Output

Accepted Modalities:

text

Performance Tips:

Apply post-training for conversational tasks; adjust `rope_scaling` for long contexts.

Release Notes

Version:

2.5

Date:

2024

Notes:

Significant improvements in code capabilities and long-context support.

LLM Name	Qwen2.5 Coder 14B
Repository 🤗	https://huggingface.co/Qwen/Qwen2.5-Coder-14B
Base Model(s)	Qwen/Qwen2.5-14B Qwen/Qwen2.5-14B
Model Size	14b
Required VRAM	29.7 GB
Updated	2026-05-20
Maintainer	Qwen
Model Type	qwen2
Model Files	5.0 GB: 1-of-6 5.0 GB: 2-of-6 5.0 GB: 3-of-6 5.0 GB: 4-of-6 5.0 GB: 5-of-6 4.7 GB: 6-of-6
Supported Languages	en
Generates Code	Yes
Model Architecture	Qwen2ForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.43.1
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	152064
Torch Data Type	bfloat16
Errors	replace

Quantized Models of the Qwen2.5 Coder 14B

Model	Likes	Downloads	VRAM
Qwen2.5 Coder 14B Bnb 4bit	6	1652	9 GB

Best Alternatives to Qwen2.5 Coder 14B

Best Alternatives	Context / RAM	Downloads	Likes
ZYH LLM Qwen2.5 14B V4	986K / 29.7 GB	36	8
Qwen2.5 14B YOYO V4	986K / 29.7 GB	13	5
ZYH LLM Qwen2.5 14B V2	986K / 29.7 GB	29	1
ZYH LLM Qwen2.5 14B V5	986K / 29.7 GB	25	8
Qwen2.5 14B YOYO V5	986K / 29.7 GB	2	4
Mergekit Model Stock Ggfjmyf	986K / 29.7 GB	14	0
Mergekit Model Stock Injkqri	986K / 29.7 GB	3	1
Primal Opus 14B Optimus V2	128K / 29.7 GB	56	4
Qwen2.5 14B ReasoningMerge	128K / 29.7 GB	9	3
Qwen2.5 14B YOYO Latest	128K / 29.7 GB	53	0

Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen2.5-Coder-14B.

Rank the Qwen2.5 Coder 14B Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

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

What open-source LLMs or SLMs are you in search of? 53972 in total.

Email us: info@extractum.io. Our Privacy Policy | Terms and Conditions | Suggest an improvement.

Our Social Media →

Original data from HuggingFace, OpenCompass and various public git repos.

Check out Ag3ntum — our secure, self-hosted AI agent for server management.

Release v20260328a

Support LLM Explorer

Qwen2.5 Coder 14B by Qwen