Dataset · Hugging Face

OpenGVLab/MMPR-Tiny

Name: MMPR-Tiny
Creator: OpenGVLab
License: mit
Keywords: task_categories:image-text-to-text, language:en, license:mit, size_categories:1M<n<10M, arxiv:2508.18265, arxiv:2411.10442, arxiv:2507.18071, region:us, multimodal, vision-language-model, reinforcement-learning, preference-data, reasoning, agentic, gui-interaction

MMPR-Tiny This is the training data used during the online RL stage of InternVL3.5, which greatly improves the overall performance of InternVL3.5 across all scales. Our training code is also open-sourced. Based on MMPR-v1.2, we compute the accuracy of each query using the provided rollouts and select those whose model accuracy falls between 0.2 and 0.8 for online RL. We further extend the dataset with recent multimodal datasets to enhance diversity. Please refer to our paper for… See the full description on the dataset page: https://huggingface.co/datasets/OpenGVLab/MMPR-Tiny.

View on Hugging FaceReport an issue

Author / organization	OpenGVLab
License (as declared)	mit
Downloads	204 (as reported by Hugging Face)
Likes	11
Created	28 August 2025
Last updated at source	31 August 2025

multimodalvision-language-modelreinforcement-learningpreference-datareasoningagenticgui-interactionembodied-agency

Embed this entry

<iframe src="https://silytics.si/embed/item/dce1b568-ef6c-4dd5-beb3-f293be7c4f2b" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Connected in the knowledge graph

OpenGVLab/InternVL3_5-4B-MPOModel · used to train
mradermacher/InternVL3_5-1B-GGUFModel · used to train
mradermacher/InternVL3_5-4B-GGUFModel · used to train
OpenGVLab/InternVL3_5-8B-PretrainedModel · used to train
OpenGVLab/InternVL3_5-GPT-OSS-20B-A4B-PreviewModel · used to train
OpenGVLab/InternVL3_5-4BModel · used to train
mradermacher/InternVL3_5-2B-i1-GGUFModel · used to train
mradermacher/InternVL3_5-30B-A3B-i1-GGUFModel · used to train
OpenGVLab/InternVL3_5-30B-A3BModel · used to train
OpenGVLab/InternVL3_5-38B-MPOModel · used to train
mradermacher/InternVL3_5-1B-i1-GGUFModel · used to train
OpenGVLab/InternVL3_5-38B-FlashModel · used to train
OpenGVLab/InternVL3_5-1B-HFModel · used to train
OpenGVLab/InternVL3_5-1BModel · used to train
mradermacher/InternVL3_5-8B-GGUFModel · used to train
OpenGVLab/InternVL3_5-2B-FlashModel · used to train
mlx-community/InternVL3_5-1B-4bitModel · used to train
OpenGVLab/InternVL3_5-14BModel · used to train
mradermacher/InternVL3_5-4B-i1-GGUFModel · used to train
mradermacher/InternVL3_5-30B-A3B-GGUFModel · used to train
OpenGVLab/InternVL3_5-8B-HFModel · used to train
OpenGVLab/InternVL3_5-8B-MPOModel · used to train
OpenGVLab/InternVL3_5-38B-InstructModel · used to train
OpenGVLab/InternVL3_5-241B-A28B-InstructModel · used to train
OpenGVLab/InternVL3_5-GPT-OSS-20B-A4B-Preview-HFModel · used to train
OpenGVLab/InternVL3_5-241B-A28B-HFModel · used to train
OpenGVLab/InternVL3_5-30B-A3B-HFModel · used to train
cyankiwi/InternVL3_5-38B-AWQ-4bitModel · used to train
OpenGVLab/InternVL3_5-8BModel · used to train
OpenGVLab/InternVL3_5-4B-HFModel · used to train
mradermacher/InternVL3_5-8B-i1-GGUFModel · used to train
cyankiwi/InternVL3_5-8B-AWQ-8bitModel · used to train
OpenGVLab/InternVL3_5-2BModel · used to train
OpenGVLab/InternVL3_5-38B-HFModel · used to train
OpenGVLab/InternVL3_5-1B-FlashModel · used to train
mradermacher/InternVL3_5-14B-i1-GGUFModel · used to train
OpenGVLab/InternVL3_5-30B-A3B-InstructModel · used to train
OpenGVLab/InternVL3_5-4B-PretrainedModel · used to train
cyankiwi/InternVL3_5-8B-AWQ-4bitModel · used to train
OpenGVLab/InternVL3_5-30B-A3B-FlashModel · used to train

Explore this entry in the graph

More from OpenGVLab

OpenGVLab/MVBenchDataset
OpenGVLab/VideoChat-Flash-Training-DataDataset
OpenGVLab/SA-Med2D-20MDataset
OpenGVLab/GUI-OdysseyDataset
OpenGVLab/ScaleCUA-DataDataset
OpenGVLab/OmniCorpus-CCDataset

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.