What are the hardware requirements for Vicuna 33B AWQ?

Vicuna 33B AWQ requires approximately 17.6 GB of VRAM and supports a context window of 2K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Vicuna 33B AWQ and how large is it?

Vicuna 33B AWQ is developed by TheBloke, a model with 33b 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 Vicuna 33B AWQ?

Vicuna 33B AWQ is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Vicuna 33B AWQ by TheBloke — VRAM 17.6GB, 2K context

Name: Vicuna 33B AWQ
Author: TheBloke

Vicuna 33B AWQ is an open-source language model by TheBloke. Features: 33b LLM, VRAM: 17.6GB, Context: 2K, License: other, Quantized, LLM Explorer Score: 0.09.

Arxiv:2302.13971 Arxiv:2306.05685 4-bit Awq Base model:lmsys/vicuna-33b-v1... Base model:quantized:lmsys/vic... Llama Quantized Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/TheBloke/vicuna-33B-AWQ

Vicuna 33B AWQ Parameters and Internals

Model Type

An auto-regressive language model based on the transformer architecture

Use Cases

Areas:

research

Applications:

large language models, chatbots

Additional Notes

AWQ is an efficient, accurate and blazing-fast low-bit weight quantization method, currently supporting 4-bit quantization. Compared to GPTQ, it offers faster Transformers-based inference.

Training Details

Data Sources:

ShareGPT

Data Volume:

125K conversations

Methodology:

Supervised instruction fine-tuning

Model Architecture:

Transformer

Input Output

Input Format:

A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. USER: {prompt} ASSISTANT:

Accepted Modalities:

text

LLM Name	Vicuna 33B AWQ
Repository 🤗	https://huggingface.co/TheBloke/vicuna-33B-AWQ
Model Name	Vicuna 33B V1.3
Model Creator	Large Model Systems Organization
Base Model(s)	lmsys/vicuna-33b-v1.3 lmsys/vicuna-33b-v1.3
Model Size	33b
Required VRAM	17.6 GB
Updated	2026-07-19
Maintainer	TheBloke
Model Type	llama
Model Files	10.0 GB: 1-of-2 7.6 GB: 2-of-2
AWQ Quantization	Yes
Quantization Type	awq
Model Architecture	LlamaForCausalLM
License	other
Context Length	2048
Model Max Length	2048
Transformers Version	4.28.1
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
Vocabulary Size	32000
Torch Data Type	float16

Rank the Vicuna 33B AWQ 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 Vicuna 33B AWQ

Best Alternatives	Context / RAM	Downloads	Likes
WhiteRabbitNeo 33B V1 AWQ	16K / 18.1 GB	802	1
Everyone Coder 33B Base AWQ	16K / 18.1 GB	8	2
WizardCoder 33B V1.1 AWQ	16K / 18.1 GB	6	5
...eepseek Coder 33B Instruct AWQ	16K / 18.1 GB	731	42
Deepseek Coder 33B Base AWQ	16K / 18.1 GB	22	4
Code 33B AWQ	2K / 17.6 GB	4	3
Python Code 33B AWQ	2K / 17.6 GB	3	1
Vicuna 33B Coder AWQ	2K / 17.6 GB	5	1
Fin Llama 33B AWQ	2K / 17.6 GB	4	5
Chronoboros 33B AWQ	2K / 17.6 GB	6	2

Note: green Score (e.g. "73.2") means that the model is better than TheBloke/vicuna-33B-AWQ.