The first On-device Agent Model Series: Deep Search Exploration and Deep Report Generation.
AI & ML interests
Large Language Models
Recent Activity
Papers
InfLLM-V2: Dense-Sparse Switchable Attention for Seamless Short-to-Long Adaptation
MiniCPM4: Ultra-Efficient LLMs on End Devices
Multimodal models with leading performance.
-
MiniCPM-V: A GPT-4V Level MLLM on Your Phone
Paper • 2408.01800 • Published • 88 -
openbmb/MiniCPM-V-4_5
Image-Text-to-Text • 9B • Updated • 57.9k • 1.05k -
openbmb/MiniCPM-V-4_5-gguf
Image-Text-to-Text • 8B • Updated • 21.4k • 49 -
openbmb/MiniCPM-V-4_5-int4
Image-Text-to-Text • 9B • Updated • 2.48k • 11
The MiniCPM family of LLMs and VLLMs.
The collection of open-source models that adopt Ultra Series datasets for training
CPM-Bee series models.
Parsing-free RAG supported by VLMs
-
VisRAG 2.0: Evidence-Guided Multi-Image Reasoning in Visual Retrieval-Augmented Generation
Paper • 2510.09733 • Published • 4 -
VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality Documents
Paper • 2410.10594 • Published • 28 -
openbmb/EVisRAG-7B
8B • Updated • 90 • 4 -
openbmb/EVisRAG-3B
4B • Updated • 48 • 1
The first On-device Agent Model Series: Deep Search Exploration and Deep Report Generation.
Multimodal models with leading performance.
-
MiniCPM-V: A GPT-4V Level MLLM on Your Phone
Paper • 2408.01800 • Published • 88 -
openbmb/MiniCPM-V-4_5
Image-Text-to-Text • 9B • Updated • 57.9k • 1.05k -
openbmb/MiniCPM-V-4_5-gguf
Image-Text-to-Text • 8B • Updated • 21.4k • 49 -
openbmb/MiniCPM-V-4_5-int4
Image-Text-to-Text • 9B • Updated • 2.48k • 11
MiniCPM4: Ultra-Efficient LLMs on End Devices
The MiniCPM family of LLMs and VLLMs.
Extrapolating RLVR to General Domains without Verifiers
The collection of open-source models that adopt Ultra Series datasets for training
UltraLM, UltraRM and UltraCM.
CPM-Bee series models.
Advancing LLM Reasoning Generalists with Preference Trees
Parsing-free RAG supported by VLMs
-
VisRAG 2.0: Evidence-Guided Multi-Image Reasoning in Visual Retrieval-Augmented Generation
Paper • 2510.09733 • Published • 4 -
VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality Documents
Paper • 2410.10594 • Published • 28 -
openbmb/EVisRAG-7B
8B • Updated • 90 • 4 -
openbmb/EVisRAG-3B
4B • Updated • 48 • 1
Embedding, re-ranking, generation -- the cornerstone of RAG.