About 20 results
Open links in new tab
  1. MiMo-V2-Flash | Xiaomi

    Dec 16, 2025 · MiMo-V2-Flash is a Mixture-of-Experts model with 309B total parameters and 15B active parameters, adopting a hybrid attention architecture that interleaves sliding-window and full attention, …

  2. Xiaomi MiMo-V2-Flash: Complete Guide to the 309B Parameter ...

    Dec 17, 2025 · The MoE architecture activates only 15B of 309B total parameters per request, combined with Multi-Token Prediction that triples generation speed, making it faster than traditional dense …

  3. MiMo-V2-Flash: 309B Open Source AI Model | Fastest LLM by Xiaomi

    Released on December 16, 2025, it features 309B total parameters with 15B active parameters, delivering 150 tokens/sec inference speed and achieving #1 on SWE-Bench Verified among open …

    • Reviews: 512
    • Xiaomi MiMo-V2-Flash: Complete Guide to the Powerful 309B ...

      MiMo-V2-Flash is Xiaomi's ambitious entry into the large language model (LLM) landscape, featuring a Mixture-of-Experts (MoE) architecture with 309 billion total parameters and only 15 billion active …

    • GitHub - XiaomiMiMo/MiMo-V2-Flash: MiMo-V2-Flash: Efficient ...

      Dec 16, 2025 · MiMo-V2-Flash is a Mixture-of-Experts (MoE) language model with 309B total parameters and 15B active parameters. Designed for high-speed reasoning and agentic workflows, it …

    • MiMo-V2-Flash: Xiaomi's 309B MoE Open-Weight Model Guide

      Dec 15, 2025 · Xiaomi has entered the frontier AI race with MiMo-V2-Flash, a 309B parameter MoE model that achieves state-of-the-art open-source performance on software engineering benchmarks …

    • MiMo‑V2‑Flash: Xiaomi’s 309B MoE Speed Demon ... - Medium

      Dec 27, 2025 · MiMo‑V2‑Flash is Xiaomi’s 309B‑parameter Mixture‑of‑Experts model with only 15B active parameters per token, built for fast reasoning, coding, and agentic workflows with 256K …