Together AI 发布了 MiniMax-M3，这是一款新的开源权重、原生多模态模型。该模型支持文本、图像和视频输入，并通过 MiniMax Sparse Attention 实现 100 万 token 的上下文窗口。它为复杂的推理和延迟敏感的聊天提供了不同的模式，与之前的版本相比速度有了显著提升。 AI
影响 此次发布提供了一个强大的开源权重多模态模型，具有大型上下文窗口，有可能加速需要跨多种数据类型进行复杂推理的领域的研究和开发。
排序理由 AI 实验室发布新模型。
在 X — Together (inference / OSS) 阅读 →
AI 生成摘要 · Google Gemini · 来自 6 个来源。 我们如何撰写摘要 →
Try it now: https://t.co/V6tQnswpBw
Highlights: 👉 Native multimodality across text, image, and video 👉 1M context with MiniMax Sparse Attention for long-context workloads 👉 9x prefill and 15x decode speedups vs M2 at 1M context 👉 Thinking mode for complex reasoning, non-thinking mode for latency-sensitive chat
MiniMax-M3 from @MiniMax_AI is now available on Together AI. It’s an open-weight native multimodal model with 1M context, MiniMax Sparse Attention, and thinking / non-thinking modes. Together AI is MiniMax’s preferred cloud partner, with inference optimizations delivering up to…
Try it now: https://t.co/V6tQnswpBw
Highlights: 👉 Native multimodality across text, image, and video 👉 1M context with MiniMax Sparse Attention for long-context workloads 👉 9x prefill and 15x decode speedups vs M2 at 1M context 👉 Thinking mode for complex reasoning, non-thinking mode for latency-sensitive chat
MiniMax-M3 from @minimax_ai is now available on Together AI. It’s an open-weight native multimodal model with 1M context, MiniMax Sparse Attention, and thinking / non-thinking modes. Together AI is MiniMax’s preferred cloud partner, with inference optimizations delivering up ht…