What are the hardware requirements for Qwen2.5 3B Haizi?

Qwen2.5 3B Haizi requires approximately 6.2 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 3B Haizi and how large is it?

Qwen2.5 3B Haizi is developed by shikunpunk, a model with 3b 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 3B Haizi?

Qwen2.5 3B Haizi is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen2.5 3B Haizi by shikunpunk — VRAM 6.2GB, 32K context

Name: Qwen2.5 3B Haizi
Author: shikunpunk

Qwen2.5 3B Haizi is an open-source language model by shikunpunk. Features: 3b LLM, VRAM: 6.2GB, Context: 32K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.29.

Base model:adapter:qwen/qwen2.... Base model:qwen/qwen2.5-3b-ins... Conversational Haizi Instruct Lora Modern-poetry Poetry-generation Qwen Qwen2 Region:us Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/shikunpunk/Qwen2.5-3B-Haizi

Qwen2.5 3B Haizi Parameters and Internals

LLM Name	Qwen2.5 3B Haizi
Repository 🤗	https://huggingface.co/shikunpunk/Qwen2.5-3B-Haizi
Base Model(s)	Qwen/Qwen2.5-3B-Instruct Qwen/Qwen2.5-3B-Instruct
Model Size	3b
Required VRAM	6.2 GB
Updated	2026-08-13
Maintainer	shikunpunk
Model Type	qwen2
Instruction-Based	Yes
Model Files	4.0 GB: 1-of-2 2.2 GB: 2-of-2
Supported Languages	zh
Model Architecture	Qwen2ForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	5.8.0
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151936
LoRA Model	Yes
Errors	replace

Rank the Qwen2.5 3B Haizi 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 Qwen2.5 3B Haizi

Best Alternatives	Context / RAM	Downloads	Likes
VibeThinker 3B	128K / 6.2 GB	52745	806
VibeThinker 3B	128K / 6.2 GB	361	0
Saba2 3B	128K / 6.2 GB	6	0
Tessa T1 3B	117K / 6.2 GB	10	5
UIGEN T1.5 3B	117K / 6.2 GB	7	1
Qwen2.5 3B Instruct	32K / 6.2 GB	5889234	545
SmallThinker 3B Preview	32K / 6.8 GB	30529	412
ResearchMate Qwen2.5 3B	32K / 6.2 GB	66	2
LeeChanRX 3B Instruct	32K / 6.2 GB	570	2
Neuron V1 3B Instruct	32K / 6.2 GB	149	0

Note: green Score (e.g. "73.2") means that the model is better than shikunpunk/Qwen2.5-3B-Haizi.