What are the hardware requirements for HPD Parsing?

HPD Parsing requires approximately 2.1 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed HPD Parsing and how large is it?

HPD Parsing is developed by PaddlePaddle, a model with 1b 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 HPD Parsing?

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

HPD Parsing by PaddlePaddle — VRAM 2.1GB

Name: HPD Parsing
Author: PaddlePaddle

HPD Parsing is an open-source language model by PaddlePaddle. Features: 1b LLM, VRAM: 2.1GB, License: apache-2.0, LLM Explorer Score: 0.4.

Arxiv:2607.18839 Base model:finetune:opengvlab/... Base model:opengvlab/internvl3... Conversational Custom code Document-parse En Eval results Feature-extraction Hpd-parsing Image-text-to-text Internvl chat P-mtp Paddleocr Region:us Safetensors Speculative-decoding Visual-language Zh

Model Card on HF 🤗: https://huggingface.co/PaddlePaddle/HPD-Parsing

HPD Parsing Parameters and Internals

LLM Name	HPD Parsing
Repository 🤗	https://huggingface.co/PaddlePaddle/HPD-Parsing
Base Model(s)	InternVL3 5 1B OpenGVLab/InternVL3_5-1B
Model Size	1b
Required VRAM	2.1 GB
Updated	2026-08-09
Maintainer	PaddlePaddle
Model Type	internvl_chat
Model Files	2.1 GB
Supported Languages	en zh
Model Architecture	InternVLChatModel
License	apache-2.0
Model Max Length	14588
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151681
Errors	replace

Rank the HPD Parsing 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 HPD Parsing

Best Alternatives	Context / RAM	Downloads	Likes
InternVL3 5 1B	0K / 2.1 GB	118776	29
InternVL3 1B	0K / 1.9 GB	420183	85
InternVL2 1B	0K / 1.9 GB	824111	82
InternVL3 5 1B Instruct	0K / 2.1 GB	1133	7
InternVL3 5 1B Flash	0K / 2.3 GB	85	11
InternVL2 5 1B	0K / 1.9 GB	20818	69
InternVL3 5 1B MPO	0K / 2.1 GB	43	2
InternVL3 5 1B Pretrained	0K / 2.1 GB	40	1
InternVL3 1B Instruct	0K / 1.9 GB	1282	9
InternVL3 1B Pretrained	0K / 1.9 GB	44	3