TAAFT
Free mode
100% free
Freemium
Free Trial
Deals

Marco-o1

By Alibaba
Marco-o1 is an open reasoning model family from Alibaba International Digital Commerce built to go beyond benchmark-style reasoning into more practical, open-ended tasks. The repository describes it as focusing not only on math, physics, and coding, but also on open-ended solutions, with the goal of becoming a general model for agentic applications that includes comprehensive planning and function-call abilities. The repo also shows the family evolving through v1, v2, v3, and DEPO releases, with v2 adding stronger self-built data and DPO-based optimization for math, planning, and instruction following, while v3 adds context compression and lower inference cost.
Text Gen 7
Released: November 13, 2024

Overview

Marco-o1 is Alibaba International Digital Commerce’s open reasoning model line aimed at real-world problem solving, not just standard-answer tasks. It is designed for mathematics, physics, coding, planning, and open-ended solutions, with an emphasis on broader agentic use, including planning ability and function-calling readiness.

About Alibaba

Chinese e-commerce and cloud leader behind Taobao, Tmall, and Alipay.

Industry: Retail
Company Size: 128197
Location: CN
Website: alibaba.com
View Company Profile

Tools using Marco-o1

No tools found for this model yet.

Last updated: April 6, 2026
0 AIs selected
Clear selection
#
Name
Task