MiniMax-M3
defaultMiniMax's flagship (Jun 2026): built for frontier coding and agents, 1M context, text, image and video input.
- Context window
- 1M tokens
- Max output
- — tokens
- Input / output
- text, image, video / text
- Latency
- slower (deliberate)
- Structured output
- no
- Tool use / native search
- yes / no
How to prompt it: Reasons by default (thinking: disabled turns it off). Temperature and top_p are fixed. In tool loops, send the model's full response back each turn, reasoning_content included.
Verified on November 14, 2026 · official source