ModelScope Agents Released in Open Source Version of GPTS by Modu Community

站长之家

Published inAI News · 2 min read · Nov 20, 2023

141

During the Openai DEV day, the ModelScope Community introduced the open-source version of GPTS, "ModelScope Agents". This tool allows users to customize and develop it further, featuring open-source availability, support for multiple LLM Agents, and the ability to finetune. Users can access and utilize this tool through Colab and Notebook links, installing and using it within a Python environment. In the future, the tool is planned to support production-level functionalities, effective evaluation methods, and usage on the ModelScope Creative Space. The ModelScope Community aims to foster the prosperity of the application ecosystem and provide similar application building experiences to users in the open-source community through this middleware.

Domestic Open Source Large Model DeepSeek R1 Released, Performance Surpasses o1 Model with 90% Cost Reduction

Recently, the DeepSeek team in China has launched its latest open-source large model R1, which has gained widespread attention. The performance of the R1 model is exceptionally outstanding, surpassing OpenAI's o1 model in multiple tests, especially excelling in assessments related to mathematics and programming. In the latest AIME2024 test in the United States, R1 scored 79.8, outperforming o1's score of 79.2. In the MATH-500 test, R1 achieved a score of 97.3, again leading over o1.

MiniMax Open Sources MiniMax-01 New Series Models, Performance Comparable to GPT-4o

On January 15, 2025, MiniMax announced the open sourcing of its new series of models, MiniMax-01, which includes the foundational language model MiniMax-Text-01 and the visual multimodal model MiniMax-VL-01. The MiniMax-01 series features bold innovations in its architecture, implementing linear attention mechanisms on a large scale for the first time, breaking the limitations of traditional Transformer architectures. With a parameter count reaching 456 billion and 45.9 billion activations per instance, its overall performance competes with international standards.

Tsinghua, Fudan, and Stanford Collaborate to Open Source the 'Eko' Framework for Automating Agent Operations on Computers

Recently, research teams from Tsinghua University, Fudan University, and Stanford University jointly released an Agent development framework called 'Eko'. This framework aims to help developers quickly build production-ready 'virtual employees' using simple code and natural language. The Eko framework can take over a user's computer and browser, performing various tedious tasks on behalf of humans. With Eko, users can achieve automated data collection, testing, and file management, among other functionalities. For instance, users can set Eko to automatically gather information from Yahoo Finance.

AI News

ModelScope Agents Released in Open Source Version of GPTS by Modu Community

站长之家

AI News Recommendations

Domestic Open Source Large Model DeepSeek R1 Released, Performance Surpasses o1 Model with 90% Cost Reduction

Byte跳豆包 Launches New AI Programming Feature, Integrates GitHub Open Source Repository

MiniMax Open Sources MiniMax-01 New Series Models, Performance Comparable to GPT-4o

Tsinghua, Fudan, and Stanford Collaborate to Open Source the 'Eko' Framework for Automating Agent Operations on Computers