MiniCPM-o-4.5-ascend-FlagOS

Introduction

FlagOS is a unified heterogeneous computing software stack for large models, co-developed with leading global chip manufacturers. With core technologies such as the FlagScale, together with vllm-plugin-fl, distributed training/inference framework, FlagGems universal operator library, FlagCX communication library, and FlagTree unified compiler, the FlagRelease platform leverages the FlagOS stack to automatically produce and release various combinations of <chip + open-source model>. This enables efficient and automated model migration across diverse chips, opening a new chapter for large model deployment and application.

Based on this, the MiniCPM-o-4.5-ascend-FlagOS model is adapted for the Hygon chip using the FlagOS software stack, enabling:

Integrated Deployment

• Deep integration with the open-source FlagScale framework
• Out-of-the-box inference scripts with pre-configured hardware and software parameters
• Released FlagOS-Ascend container image supporting deployment within minutes

Consistency Validation

• Rigorously evaluated through benchmark testing: Performance and results from the FlagOS software stack are compared against native stacks on multiple public.

Technical Overview

FlagGems

FlagGems is a high-performance, generic operator library implemented in Triton language. It is built on a collection of backend-neutral kernels that aims to accelerate LLM (Large-Language Models) training and inference across diverse hardware platforms.

FlagTree

FlagTree is an open source, unified compiler for multiple AI chips project dedicated to developing a diverse ecosystem of AI chip compilers and related tooling platforms, thereby fostering and strengthening the upstream and downstream Triton ecosystem. Currently in its initial phase, the project aims to maintain compatibility with existing adaptation solutions while unifying the codebase to rapidly implement single-repository multi-backend support. For upstream model users, it provides unified compilation capabilities across multiple backends; for downstream chip manufacturers, it offers examples of Triton ecosystem integration.

FlagScale

FlagScale is a comprehensive toolkit designed to support the entire lifecycle of large models. It builds on the strengths of several prominent open-source projects, including Megatron-LM and vLLM, to provide a robust, end-to-end solution for managing and scaling large models.

FlagCX

FlagCX is a scalable and adaptive cross-chip communication library. It serves as a platform where developers, researchers, and AI engineers can collaborate on various projects, contribute to the development of cutting-edge AI solutions, and share their work with the global community.

FlagEval Evaluation Framework

FlagEval is a comprehensive evaluation system and open platform for large models launched in 2023. It aims to establish scientific, fair, and open benchmarks, methodologies, and tools to help researchers assess model and training algorithm performance. It features:

• Multi-dimensional Evaluation: Supports 800+ model evaluations across NLP, CV, Audio, and Multimodal fields, covering 20+ downstream tasks including language understanding and image-text generation.
• Industry-Grade Use Cases: Has completed horizontal evaluations of mainstream large models, providing authoritative benchmarks for chip-model performance validation.

Evaluation Results

In the following table, +x means FlagOS performs better than CUDA with x accuracy score, while -x means CUDA performs better. For FlagEval's benchmarks avg@1, +-5.00 means there's no significant difference between two AI-ecosystems.

Transformers version

Accuracy Difference between using FLAGOS and Ascend on Nvidia-CUDA

User Guide

Environment Setup

Operation Steps

Download FlagOS Image

docker pull  harbor.baai.ac.cn/flagrelease-public/flagrelease-ascend-release-model_minicpm-o-45-tree_none-gems_4.2.0-scale_none-cx_none-python_3.11.13-torch_npu2.8.0-pcp_cann8.3.0.2.220_8.3.rc2-gpu_ascend001-arc_arm64-driver_25.2.0:latest

Download Open-source Model Weights

pip install modelscope
modelscope download --model FlagRelease/MiniCPM-o-4.5-ascend-FlagOS --local_dir /share/MiniCPMO45

Start the Container

#Container Startup
docker run -itd --name=ascend --privileged --ipc=host --network=host --shm-size=500G -w /workspace/cpmo -v /data/workspace-video-mme/video-mme:/data -v /usr/local/Ascend/driver:/usr/local/Ascend/driver -v /usr/local/Ascend/firmware:/usr/local/Ascend/firmware -v /usr/local/sbin/npu-smi:/usr/local/sbin/npu-smi -v /usr/local/dcmi:/usr/local/dcmi -v /usr/local/sbin:/usr/local/sbin -v /usr/bin/hostname:/usr/bin/hostname -v /etc/ascend_install.info:/etc/ascend_install.info -v /var/log/npu/:/usr/slog -v /etc/hccn.conf:/etc/hccn.conf -v /etc/localtime:/etc/localtime -v /etc/hosts:/etc/hosts -v /root/.cache:/root/.cache -v /share:/share harbor.baai.ac.cn/flagrelease-public/flagrelease-ascend-release-model_minicpm-o-45-tree_none-gems_4.2.0-scale_none-cx_none-python_3.11.13-torch_npu2.8.0-pcp_cann8.3.0.2.220_8.3.rc2-gpu_ascend001-arc_arm64-driver_25.2.0:latest
docker exec -it ascend bash

User Reference

MiniCPM-o-4.5 is a model that supports Omni-input and offers a wide range of usage scenarios. FlagRelease only provides migration results of the FlagOS software ecosystem across various chips and does not impose any restrictions on how users choose to utilize the model. You are encouraged to refer to the official OpenBMB README for flexible and comprehensive usage guidance. That said, we do provide a basic example: performing video-based question answering with spoken audio output.

cd /root/code
ln -s /share/MiniCPMO45 ./MiniCPMO45
python openbmb_main.py

Contributing

We warmly welcome global developers to join us:

1. Submit Issues to report problems
2. Create Pull Requests to contribute code
3. Improve technical documentation
4. Expand hardware adaptation support

License

The weight files are from MiniCPM-o-4.5 from OpenBMB, open source with apache2.0 licensehttps://www.apache.org/licenses/LICENSE-2.0.txt.