multimodal LLMfamily
LLaVA
Visual instruction tuning: a vision encoder connected to an open LLM and trained on GPT-generated image instruction data, the template for many open VLMs.
Levelfamily
Classmultimodal LLM
Parametersnot in the notes
Contextnot in the notes
Released17 Apr 2023
Opennessunknown
Runs onnot in the notes
Statusactive
Lineage
LLaVA-OneVision Qwen2 7B belongs to this familymultimodal LLM13 Aug 20248.03BLLaVA-NeXT-Video 7B belongs to this familymultimodal LLM5 Jun 20247.06BLLaVA-NeXT (1.6) Mistral 7B belongs to this familymultimodal LLM20 Feb 20247.57BLLaVA-1.5 7B belongs to this familymultimodal LLM5 Dec 20237.06BWhere 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
Identity
- Developer
- University of Wisconsin-Madison, Microsoft AI
- Family
- not in the notes
- Released
- 17 Apr 2023 sourcesannouncement page
- 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
- –
- Note
- LLaVA, LLaVA-1.5, LLaVA-NeXT (1.6), NeXT-Video and OneVision; later versions involve ByteDance.
- Context length
- not in the notes
- Input
- text, image
- 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
- not in the notes
- 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: en, zh
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.
Links
ModelLLaVAUniversity of Wisconsin-Madison, multimodal LLM
LLaVAUniversity of Wisconsin-Madison, multimodal LLMR&D TopicsVisual Instruction Tuning
Use CasesRobot Inspection with Instrument ReadingIndustrialPathology Whole-Slide ImagingHealthcareUse cases of models in the same R&D topic; none is recorded for this model itself.
Who Runs Itnone recordedDeployments of these use cases; none names this model.
Models 4
LLaVA-NeXT-Video 7BUniversity of Wisconsin-Madison, Microsoft AI, multimodal LLMFamilyLLaVA-1.5 7BUniversity of Wisconsin-Madison, Microsoft AI, multimodal LLMFamilyLLaVA-NeXT (1.6) Mistral 7BUniversity of Wisconsin-Madison, Microsoft AI, multimodal LLMFamilyLLaVA-OneVision Qwen2 7BByteDance, multimodal LLMFamilySources
- arxiv.org/abs/2304.08485 https://arxiv.org/abs/2304.08485
- llava-vl.github.io https://llava-vl.github.io/
- github.com/haotian-liu/LLaVA https://github.com/haotian-liu/LLaVA
- github.com/GoogleCloudPlatform/vertex-ai-samples/tree/ma... https://github.com/GoogleCloudPlatform/vertex-ai-samples/tree/main/notebooks/community/model_garden
- github.com/open-mmlab/mmpretrain/tree/main/configs/llava https://github.com/open-mmlab/mmpretrain/tree/main/configs/llava
- huggingface.co/llava-hf/llava-1.5-7b-hf https://huggingface.co/llava-hf/llava-1.5-7b-hf
- huggingface.co/llava-hf/llava-v1.6-mistral-7b-hf https://huggingface.co/llava-hf/llava-v1.6-mistral-7b-hf
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "LLaVA", https://aurion.semai.id/models/llava-llava/, records of October 2026.