MiniMax-M2.5

MiniMax-M2.5 is a coding model from MiniMax for agent workflows, office edits, and automation, according to the maker. It is the latest in the minimax line, following MiniMax-M2.1. It has a 204800-token context window and a 131072-token output limit. It accepts and returns text, supports reasoning mode and tool calling, and has open weights. Pricing starts at $0.15 per million input tokens and $0.88 per million output tokens, with cached input at $0.027 per million tokens.

Against MiniMax-M2.1: input $0.27 to $0.15, output $1.10 to $0.88, cached input $0.03 to $0.027.

Range across the hosts that serve it: up to $0.60 per 1M input tokens.

• ·204800-token context window
• ·131072-token max output
• ·Supports reasoning mode
• ·Supports tool calling
• ·Open weights available

• Reach for it for agent workflows
• Reach for it for office edits
• Reach for it for automation tasks

42 hosts serve this model at the price above · the maker's documentation

The registry keeps the six most recent versions of a line. Older ones stay in the file and can be surfaced without a rebuild.

What is the context window size?

The context window is 204800 tokens, and the maximum output is 131072 tokens.

Does it support tool calling?

Yes, it supports tool calling and reasoning mode. It accepts and returns text only.

What is the pricing?

Input costs from $0.15 per million tokens, output from $0.88 per million tokens. Cached input is $0.027 per million tokens.