What are the hardware requirements for HoloCT 4B?

HoloCT 4B requires approximately 9.4 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed HoloCT 4B and how large is it?

HoloCT 4B is developed by Alibaba-DAMO-Academy, a model with 4b 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 HoloCT 4B?

HoloCT 4B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

HoloCT 4B by Alibaba-DAMO-Academy — VRAM 9.4GB

Name: HoloCT 4B
Author: Alibaba-DAMO-Academy

HoloCT 4B is an open-source language model by Alibaba-DAMO-Academy. Features: 4b LLM, VRAM: 9.4GB, License: mit.

Arxiv:2611.06471 3d Computed-tomography Conversational En Grounding Image-segmentation Image-text-to-text Medical-imaging Qwen3 vl Region:us Report-generation Safetensors Sharded Tensorflow Visual-question-answering

Model Card on HF 🤗: https://huggingface.co/Alibaba-DAMO-Academy/HoloCT-4B

HoloCT 4B Parameters and Internals

LLM Name	HoloCT 4B
Repository 🤗	https://huggingface.co/Alibaba-DAMO-Academy/HoloCT-4B
Model Size	4b
Required VRAM	9.4 GB
Updated	2026-11-05
Maintainer	Alibaba-DAMO-Academy
Model Type	qwen3_vl
Model Files	5.0 GB: 1-of-2 4.4 GB: 2-of-2
Supported Languages	en
Model Architecture	HoloCTForConditionalGeneration
License	mit
Model Max Length	262144
Transformers Version	4.57.0
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Errors	replace

Rank the HoloCT 4B 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