モデルを比較

This is the free OpenRouter endpoint for MiniMax's M3, a 428B-parameter mixture-of-experts model (23B active, MiniMax Community License, 1M-token context) with native image and video multimodality and MiniMax Sparse Attention for long-context efficiency, released mid-2026. Free access is rate-limited, making it best suited for experimenting with its coding and long-horizon agentic capabilities rather than production workloads.

GPT-6 Luna is the fast, low-cost tier of OpenAI's GPT-6 series (1.05M-token context, text, image, and file input, $0.10/$0.50 per million input/output tokens), released September 22, 2026 below GPT-6 Sol and the flagship GPT-6 Astra. It is built for high-volume tasks with a clear goal, such as summarization, extraction, classification, quick questions, and lightweight agentic steps.