openbmb

Multi-level memory plugin for OpenClaw that adds long-term context across agent sessions.

Router for fanning out Claude-style requests across multiple LLM backends — drop-in for Claude Code and other Anthropic-API clients.

Benchmark and evaluation toolkit for retrieval-augmented generation pipelines — measure how well your RAG actually answers questions.

Reference iOS/Swift apps showing how to run MiniCPM-V multimodal models on-device — drop-in starting point for vision apps.

Open-source framework for evaluating foundation models — modular benchmarks for LLM capability assessment (ACL 2024 Demo).

General-purpose mobile agent driven by multimodal foundation models for long-horizon tasks on phone apps.

Open-source framework for evaluating audio and speech models across recognition, speech-to-text, and speech-to-speech tasks.

OpenBMB's reasoning-augmented vision-language-action model for robotics, combining reinforcement learning with explicit thinking.

Open-source framework for collaborative AI agents. Distributed agents discover each other and team up over an internet-like protocol.

End-to-end infrastructure for training and evaluating LLM agents. Tooling for the full agent lifecycle from data to eval.

LLM-powered agent that explores a repo and auto-generates documentation so teams can onboard faster.

Parsing-free RAG pipeline backed by vision-language models — index PDFs and slides as images instead of text.

OpenBMB's low-code MCP framework for building complex RAG pipelines — drag-and-drop retrieval, rerank, and generation components.

OpenBMB's open platform for training, serving, and evaluating LLMs for tool learning — over 16K real-world APIs in the training set.

OpenBMB's framework for deploying multiple LLM-based agents across applications — simulation, task solving, and multi-agent collaboration.

OpenBMB's family of ultra-efficient end-side LLMs (MiniCPM 4 and 4.1) that run on phones and edge devices with strong reasoning per parameter.

Pocket-sized multimodal LLM from OpenBMB designed for efficient image and video understanding on phones and edge devices.

A tokenizer-free text-to-speech system that generates high-quality multilingual speech with voice cloning and creative voice design capabilities. Supports 30 languages and can create new voices from text descriptions alone or clone existing voices from audio samples.

A zero-code multi-agent orchestration platform that lets users build and execute customized multi-agent systems through simple configuration. Users can define agents, workflows, and tasks to orchestrate complex scenarios like data visualization, 3D generation, and deep research without any coding required.