What are the hardware requirements for InternVL2 5 38B AWQ?

InternVL2 5 38B AWQ requires approximately 30.4 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed InternVL2 5 38B AWQ and how large is it?

InternVL2 5 38B AWQ is developed by OpenGVLab, a model with 38b 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 InternVL2 5 38B AWQ?

InternVL2 5 38B AWQ is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

InternVL2 5 38B AWQ by OpenGVLab — VRAM 30.4GB

Name: InternVL2 5 38B AWQ
Author: OpenGVLab

InternVL2 5 38B AWQ is an open-source language model by OpenGVLab. Features: 38b LLM, VRAM: 30.4GB, License: mit, Quantized, LLM Explorer Score: 0.16.

Arxiv:2312.14238 Arxiv:2404.16821 Arxiv:2410.16261 Arxiv:2412.05271 Awq Base model:opengvlab/internvl2... Base model:quantized:opengvlab... Conversational Custom code Feature-extraction Image-text-to-text Internvl Internvl chat Multilingual Pytorch Quantized Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/OpenGVLab/InternVL2_5-38B-AWQ

InternVL2 5 38B AWQ Parameters and Internals

LLM Name	InternVL2 5 38B AWQ
Repository 🤗	https://huggingface.co/OpenGVLab/InternVL2_5-38B-AWQ
Base Model(s)	InternVL2 5 38B OpenGVLab/InternVL2_5-38B
Model Size	38b
Required VRAM	30.4 GB
Updated	2026-08-08
Maintainer	OpenGVLab
Model Type	internvl_chat
Model Files	2.0 GB: 1-of-16 2.0 GB: 2-of-16 2.0 GB: 3-of-16 2.0 GB: 4-of-16 2.0 GB: 5-of-16 1.2 GB: 6-of-16 2.0 GB: 7-of-16 1.9 GB: 8-of-16 1.9 GB: 9-of-16 2.0 GB: 10-of-16 1.9 GB: 11-of-16 1.9 GB: 12-of-16 1.9 GB: 13-of-16 2.0 GB: 14-of-16 1.9 GB: 15-of-16 1.8 GB: 16-of-16
AWQ Quantization	Yes
Quantization Type	awq
Model Architecture	InternVLChatModel
License	mit
Model Max Length	16384
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Torch Data Type	float16
Errors	replace

Rank the InternVL2 5 38B 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 InternVL2 5 38B AWQ

Best Alternatives	Context / RAM	Downloads	Likes
InternVL3 38B AWQ	0K / 30.4 GB	2891	4
InternVL2 5 38B MPO AWQ	0K / 30.4 GB	32	6
InternVL3 5 38B	0K / 76.9 GB	22305	45
InternVL3 5 38B Instruct	0K / 77 GB	15340	5
InternVL3 5 38B Flash	0K / 79.4 GB	106	8
InternVL3 5 38B MPO	0K / 77 GB	33	2
InternVL2 5 38B MPO	0K / 77.1 GB	33	20

Note: green Score (e.g. "73.2") means that the model is better than OpenGVLab/InternVL2_5-38B-AWQ.