Multimodal Vision-Language Model Generation

2 repos

Large-scale multimodal models that combine vision and language capabilities for image understanding and generation tasks. The cluster centers on model variants across different parameter scales (2B through 8x7B) optimized for inference deployment, with emphasis on safe tensor serialization and endpoint compatibility for production use. Repositories here represent both the core model architectures and infrastructure for serving vision-language generation at scale.

Python · 2

large-language-models ·3,513

vision-language-model ·3,513

generation ·3,324

large-multimodal-models ·189

llava ·189

llm ·189

multimodal ·189

multi-modality ·189

large-context ·189

Related clusters

Large Language Models and Multimodal AI29 LLM and Vision-Language Model Evaluation18 Vision-Language Models & Multimodal AI16 Multimodal LLM Fine-tuning & Vision-Language Models12 Multimodal Vision-Language Models19 Mistral LoRA Multimodal Fine-tuning11

dvlab-research/MGM

Official repo for "Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models"

Python

3,324

35 commits

sshh12/multi_token

Embed arbitrary modalities (images, audio, documents, etc) into large language models.

Python

189

84 commits