MiniMax: MiniMax-01

MiniMax-01 is a text-and-image input model from MiniMax with an unusually large context window of roughly one million tokens, and an equally large maximum output length. It does not support tool use, reasoning modes, or structured output, so it is best suited for long-form reading, summarization, or generation tasks rather than agentic pipelines or reliably formatted data extraction. At $0.20 per million input tokens and $1.10 per million output tokens, the pricing is competitive, particularly for workloads that lean heavily on input rather than generation. The catch is that there is no independent benchmark coverage to validate quality claims, which makes it genuinely difficult to compare against tested alternatives. Teams comfortable running their own evaluations on a low-cost, high-context model may find it worth piloting, but buyers who need verified performance data before committing should wait for third-party results.