ZGCM-1-7B model opens full weights, data, and code
ZGCM-1-7B Opens Full Weights, Data, and Code for Math and Agentic Search
This is a significant release for the open-source AI community, providing a large model for math and search tasks.
The ZGCM-1-7B model from the Chinese Academy of Sciences has been fully open-sourced, including its weights, training data, and code. It achieved about 97% on the MATH-500 math test, about 75% on the AIME 2026 test, and about 63% on the WebWalkerQA test.
ZGCM-1-7B Opens Full Weights, Data, and Code for Math and Agentic Search
Zhongguancun Academy’s ZGCM-1-7B posts ~97% on MATH-500, ~75% on AIME 2026, and ~63% on WebWalkerQA while releasing weights, data, and training code.
This report is categorized under AITOP's "Models" desk, focusing on China's AI industry and domestic foundation model ecosystem. Within our cross-regional analytical framework, it directly maps to US-China foundation model benchmarks and commercialization comparison and AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include ZGCM-1-7B, open-source models, math models, agentic search.
Primary source verification: published directly by pandaily with first-party authority.
