PulseAugur
EN
LIVE 16:57:29

MiniMax M3 launches with 1M-token context and MSA architecture

By PulseAugur Editorial · [8 sources] ·

MiniMax has released its M3 model, featuring a novel Sparse Attention (MSA) architecture that enables a 1 million token context window and native multimodality. This new architecture significantly reduces computational costs for long contexts, making M3 substantially faster than previous generations. The model also demonstrates strong performance in coding and agentic tasks, surpassing several leading models on benchmarks like SWE-Bench Pro and Terminal-Bench. AI

IMPACT Sets new SOTA on coding benchmarks and offers unprecedented context length, potentially shifting industry standards for model efficiency and capability.

RANK_REASON Frontier-lab model release with system card and new architecture.

Read on Together AI blog →

AI-generated summary · Google Gemini · from 8 sources. How we write summaries →

MiniMax M3 launches with 1M-token context and MSA architecture

COVERAGE [8]

  1. X — Fireworks (inference infra) TIER_1 English(EN) · FireworksAI_HQ ·

    MiniMax M3 arrives with MiniMax Sparse Attention (MSA), 15.6x faster decoding at 1M tokens. We're partnering with @MiniMax_AI to power the inference behind this

    MiniMax M3 arrives with MiniMax Sparse Attention (MSA), 15.6x faster decoding at 1M tokens. We're partnering with @MiniMax_AI to power the inference behind this week's launch. Head to https://t.co/zwLs8Pj7I6 to take it for a spin. Once the model weights are released, M3 will be …

  2. X — Together (inference / OSS) TIER_1 (AF) · togethercompute ·

    Speakers:

    Speakers: Pengyu Zhao, Head of Research at MiniMax Haohai Sun, Research Scientist at MiniMax Ce Zhang, Founder/CTO at Together Yineng Zhang, Senior Director at Together Dan Fu, VP of Kernels at Together Hosted by Zain Hasan, Staff AI Engineer at Together

  3. X — Together (inference / OSS) TIER_1 English(EN) · togethercompute ·

    We'll get into @MiniMax_AI M3's model performance, the MSA architecture and what it means for long context, and how Together is optimizing inference and KV-cach

    We'll get into @MiniMax_AI M3's model performance, the MSA architecture and what it means for long context, and how Together is optimizing inference and KV-cache for this new architecture. Set your reminders!

  4. X — Together (inference / OSS) TIER_1 (AF) · togethercompute ·

    Speakers:

    Speakers: Pengyu Zhao, Head of Research at MiniMax Haohai Sun, Research Scientist at MiniMax Ce Zhang, Founder/CTO at Together Dan Fu, VP of Kernels at Together Hosted by Zain Hasan, Staff AI Engineer at Together

  5. X — Together (inference / OSS) TIER_1 English(EN) · togethercompute ·

    We'll get into @MiniMax_AI M3's model performance, the MSA architecture and what it means for long context, and how Together is optimizing inference and KV-cac

    We'll get into @MiniMax_AI M3's model performance, the MSA architecture and what it means for long context, and how Together is optimizing inference and KV-cache for this new architecture. Set your reminders.

  6. X — Together (inference / OSS) TIER_1 English(EN) · togethercompute ·

    MiniMax M3 is live and Together AI is powering its inference 🚀

    MiniMax M3 is live and Together AI is powering its inference 🚀 Tomorrow at 6pm PT we're going live on X Spaces with the teams behind the model and the infrastructure to give you a deep dive. https://t.co/wPayfOWmNg

  7. Together AI blog TIER_1 English(EN) ·

    Serving MiniMax-M3 for efficient inference: Unlocking 1M-Token Context and Multimodality Without Regrets

    How Together served MiniMax-M3 efficiently with KV-block-major sparse attention, paged MSA decode, optimized index scoring, and a Rust-based multimodal gateway.

  8. MarkTechPost TIER_1 English(EN) · Asif Razzaq ·

    MiniMax Releases MiniMax M3 with MSA Architecture Supporting 1M-Token Context, Native Multimodality, and Agentic Coding

    <p>MiniMax M3 introduces MiniMax Sparse Attention, a 1M-token context window, and native image, video, and computer use support.</p> <p>The post <a href="https://www.marktechpost.com/2026/06/01/minimax-releases-minimax-m3-with-msa-architecture-supporting-1m-token-context-native-m…

RELATED ENTITIES

RELATED TOPICS