F24 SALES / MODEL

← Back to the directory

MiniMaxπŸ‡¨πŸ‡³

MiniMax M3

MiniMax's multimodal model combines coding and extended tool-based tasks with text, image and video understanding. Its sparse attention mechanism is designed to reduce computation and memory requirements when processing long inputs.

CONTEXT1M
MAX OUTPUT1M
LAST SCAN13 Nov 2026
Open weights Text

Free aggregator routes

These are the aggregator routes recorded in the current or last published catalog. Each route keeps its exact model ID and thinking information.

Limits above: Intern Discovery. Available gateway routes are listed below.

The supplied Discovery minimax-m3 ID matches the exact MiniMax release documented at https://huggingface.co/MiniMaxAI/MiniMax-M3 and in MiniMax's official model catalog. The scan's reported_provider value tokenplan does not establish model authorship. No pinned serving revision is established.

Intern Discovery
Info
Thinking
Not documented
Model ID Β· Intern Discovery APIminimax-m3

Model sources

Use it through LiteLLM

Import the current LiteLLM-compatible YAML and use the exact model ID shown above. The catalog is generated from the same reviewed scan data as the download.

Download LiteLLM YAML

Sources & references