Project AURIONAI Use-Cases Research & Innovation
multimodal LLMfamily

MiniCPM-V

Efficient open vision language models sized for phones and PCs, for OCR, document and video understanding; MiniCPM-V 4.6 is already an AURION record.

OpenBMB (Tsinghua NLP Lab and ModelBest)OpenBMB (Tsinghua NLP Lab and ModelBest)China
OpenBMB (Tsinghua NLP Lab and ModelBest)
Levelfamily
Classmultimodal LLM
Parametersnot in the notes
Contextnot in the notes
Released1 Feb 2024
Opennessunknown
Runs onconsumer GPU, mobile or NPU
Statusactive

Lineage

202420252026MiniCPM-Llama3-V 2.5MiniCPM-Llama3-V 2.5, May 2024MiniCPM-V 2.6MiniCPM-V 2.6, Aug 2024MiniCPM-V 4.5MiniCPM-V 4.5, Aug 2025
OpenBMB (Tsinghua NLP Lab and ModelBest)MiniCPM-V 4.5 belongs to this familymultimodal LLM24 Aug 20258.7B
OpenBMB (Tsinghua NLP Lab and ModelBest)MiniCPM-V 2.6 belongs to this familymultimodal LLM4 Aug 20248.1B
OpenBMB (Tsinghua NLP Lab and ModelBest)MiniCPM-Llama3-V 2.5 belongs to this familymultimodal LLM19 May 20248.54B
Where this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. Confidence and evidence are the notes' own.

R&D Topics

Document, OCR and chart understandingexemplarVisual instruction tuningproduced by

Identity

Developer
OpenBMB (Tsinghua NLP Lab and ModelBest)
Family
not in the notes
Released
1 Feb 2024 sourcesregistry date
Weights published
–
Record checked
12 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
dense transformer
Total parameters
– sources
Active parameters
–
Checkpoint parameters
–
Context length
not in the notes
Input
text, image, video
Output
text
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
unknown
Licence
not in the notes sources
Access routes
open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
consumer GPU, mobile or NPU
Footprint
not in the notes
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
not stated
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
Model card languages: multilingual
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Benchmark Results

The notes record no benchmark result for this model.

Industry Use Cases

No industry use case links this model.

Sources

  1. arxiv.org/abs/2408.01800 https://arxiv.org/abs/2408.01800
  2. github.com/OpenBMB/MiniCPM-o https://github.com/OpenBMB/MiniCPM-o
  3. huggingface.co/api/models?pipeline_tag=visual-question-answe... https://huggingface.co/api/models?pipeline_tag=visual-question-answering&sort=downloads&limit=50
  4. modelscope.cn/models/OpenBMB/MiniCPM-Llama3-V-2_5-int4 https://www.modelscope.cn/models/OpenBMB/MiniCPM-Llama3-V-2_5-int4
  5. huggingface.co/openbmb/MiniCPM-Llama3-V-2_5 https://huggingface.co/openbmb/MiniCPM-Llama3-V-2_5
  6. huggingface.co/openbmb/MiniCPM-V-2_6 https://huggingface.co/openbmb/MiniCPM-V-2_6
  7. github.com/OpenBMB/MiniCPM-V https://github.com/OpenBMB/MiniCPM-V

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "MiniCPM-V", https://aurion.semai.id/models/openbmb-minicpm-v/, records of October 2026.