'%3e%3crect%20x='16'%20y='51'%20width='68'%20height='27'%20rx='13.5'/%3e%3ccircle%20cx='32'%20cy='53'%20r='15'/%3e%3ccircle%20cx='53'%20cy='43'%20r='20'/%3e%3ccircle%20cx='71'%20cy='57'%20r='12'/%3e%3c/g%3e%3c/svg%3e){kind=small}
Cloud ChangelogMiniMax models
29 announcements since January 2025 · 3 GA
Made by MiniMax
Free. Unsubscribe any time.
November 20263
MiniMax M3.1 Flash adapters can be deployed on dedicated model inference
Together AI · MiniMax-M3.1-Flash
MiniMax-M3 deprecated for new post-training resources, use MiniMax-M3.1-Flash
Together AI · MiniMax-M3 · MiniMax-M3.1-Flash
October 20264
MiniMax-M3.1-Flash available for LoRA post-training with 262,144 context
Together AI · MiniMax-M3.1-Flash
MiniMax-M3.1-Flash on serverless with 1,048,576-token context window
Together AI · MiniMax-M3.1-Flash
Six Serverless models deprecated, to be decommissioned on November 7, 2026
Fireworks AI · MiniMax-M3
September 20261
MiniMax-H3 available on serverless at $0.1391/sec at 2k
Together AI · MiniMax-H3
August 20261
MiniMax M2.7, GPT OSS 20B and three more models deprecated from serverless
Fireworks AI · MiniMax-M2.7
July 20263
MiniMax-M2.7 deprecated and removed from serverless
Together AI · MiniMax-M2.7
Provisioned throughput available for MiniMax-M3 and GLM-5.2, one-month minimum term
Together AI · MiniMax-M3
June 20264
MiniMax M2.5 deprecated from serverless, migrate to MiniMax M2.7
Fireworks AI · MiniMax M2.5 · MiniMax-M2.7
Kimi K2.7 Code, MiniMax M3 and Qwen 3.7 Plus added to the Model Library
Fireworks AI · MiniMax-M3
MiniMax-M3 on serverless with 524,288-token context, $0.30 input per 1M tokens
Together AI · MiniMax-M3
MiniMax-M3 released for agentic reasoning, tool use, coding and long context
MiniMax · MiniMax-M3
May 20261
MiniMax-M2.5 deprecated and removed from serverless
Together AI · MiniMax M2.5
April 20261
MiniMax-M2.7 now available on serverless
Together AI · MiniMax-M2.7
March 20263
Minimax M2.5 and GLM 5 models now available on Amazon BedrockGA
AWS · MiniMax M2.5
MiniMax-M2.7 and M2.7-highspeed models released
MiniMax · MiniMax-M2.7 · MiniMax-M2.7-highspeed
MiniMax-M2.5 cached input tokens priced at $0.06 per 1M, 80% off
Together AI · MiniMax M2.5
February 20264
MiniMax-M2.5 now available on serverless
Together AI · MiniMax M2.5
Amazon Bedrock adds support for six fully-managed open weights models
AWS · MiniMax M2.1
MiniMax-M2.5 and M2.5-highspeed released, with new SOTA benchmark results
MiniMax · MiniMax M2.5 · MiniMax-M2.5-highspeed
December 20252
MiniMax-M2 is now available on Amazon SageMaker JumpStartGA
AWS · MiniMax-M2
MiniMax-M2.1 released, with multi-language programming and code refactoring
MiniMax · MiniMax M2.1
October 20251
MiniMax-M2 released, with free API calls until November 7, 2025
MiniMax · MiniMax-M2
January 20251
MiniMax-Text-01 LLM and MiniMax-VL-01 vision-language model released
MiniMax · MiniMax-Text-01 · MiniMax-VL-01
All model changelogs· Jan 15, 2025 to Nov 10, 2026