Qwen3.8-Max tops CommerceAgentBench among open-weight models

Alibaba reports that Qwen3.8-Max delivers the strongest overall performance among open-weight models on CommerceAgentBench, a benchmark built from real commercial demand.

This update bears on 8 kinds of work in 6 markets. Highest level accepted: L2 Partial automation.

Highest accepted

L2Partial automation

Work

8

Markets

6

Views

73.7K

The work it bears on

Highest level first. Open a kind of work to see everything else that moved it.

• L2Partial automation
  Assess practical performance and occupational competenceVocational skills and practical training
  L2 Partial automationClaimed L3 · Accepted L2From the publisher only
• L2Partial automation
  Implement tool use and agent workflowsModel applications and agent systems
  L2 Partial automationFrom the publisher only
• L2Partial automation
  Handle product inquiries and online after-sales serviceOnline storefronts and merchandise operations
  L2 Partial automationFrom the publisher only
• L2Partial automation
  Match buyers and sellers and verify trade termsResale and trade intermediation
  L2 Partial automationFrom the publisher only
• L2Partial automation
  Assist shoppers process checkout and handle store returnsSupermarkets convenience and store operations
  L2 Partial automationFrom the publisher only
• L2Partial automation
  Request quotations negotiate and manage purchase ordersWholesale sourcing and supplier coordination
  L2 Partial automationFrom the publisher only
• L1Assisted
  Create product information and listingsOnline storefronts and merchandise operations
  L1 AssistedFrom the publisher only
• L1Assisted
  Process online orders and fulfillment exceptionsOnline storefronts and merchandise operations
  L1 AssistedFrom the publisher only

Published at