Together AI 已宣布推出 MiniMax M3 API，这是一个开放权重、原生多模态模型。该模型拥有 100 万 token 的上下文窗口，通过 MiniMax Sparse Attention 增强，并具有独特的“思考”和“非思考”模式，分别用于复杂推理和延迟敏感任务。Together AI 被强调为 MiniMax 的首选云合作伙伴，为 M3 模型提供优化的推理。 AI
影响 此次发布为开发者提供了一个强大的多模态模型，具有大上下文窗口和优化的推理能力，有可能加速需要复杂推理和处理各种数据类型的领域的开发。
排序理由 该集群宣布了一款具有 100 万上下文窗口和稀疏注意力等高级功能的新多模态模型的可用性，这属于研究和产品发布类别。
在 X — Together (inference / OSS) 阅读 →
AI 生成摘要 · Google Gemini · 来自 8 个来源。 我们如何撰写摘要 →
Highlights: 👉 Stronger real-world coding performance over Kimi K2.6 👉 ~30% lower thinking-token usage for better token efficiency 👉 256K context for larger repos and longer coding sessions 👉 Interleaved thinking and multi-step tool calling for coding agents
Try it now: https://t.co/5w9OrtVrvx
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…
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
Try it now: https://t.co/V6tQnswpBw
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…