MiniMax releases M3.1 Flash with benchmark scores and API pricing

MiniMax announces MiniMax-M3.1-Flash, an earlier version of which was Space Bunny. It reports Automation-Bench 62%, ALE 30%, DeepSWE 74%, BabyVision 91%, at 1/3 the cost of M3 and 3x the speed, with flat API pricing of $0.15/$0.60 per 1M input/output tokens and weights arriving Oct 23.

This update bears on 6 kinds of work in 5 markets. Highest level accepted: L2 Partial automation.

Highest accepted

L2Partial automation

Work

6

Markets

5

Views

701K

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 automationFrom the publisher only
• L2Partial automation
  Build research prototypes and run comparative experimentsComputing and engineering research
  L2 Partial automationFrom the publisher only
• L1Assisted
  Develop business logic and service APIsApplication interfaces and business development
  L1 AssistedFrom the publisher only
• L1Assisted
  Implement tool use and agent workflowsModel applications and agent systems
  L1 AssistedFrom the publisher only
• L1Assisted
  Reproduce diagnose and verify defect fixesSoftware testing and defect analysis
  L1 AssistedFrom the publisher only
• L1Assisted
  Run automated and exploratory testsSoftware testing and defect analysis
  L1 AssistedFrom the publisher only

Published at