MiMo-V3 is getting a new architecture. The core of it, HySparse2, is out today. Less prefill, a smaller KV cache, better long-context retrieval—and we got all three at once. Compared with MiMo-V2.6's Hybrid SWA architecture: • 5.02× lower prefill FLOPs at 1M tokens • 4.5× smaller KV cache at 1M tokens • Better MRCRv2 and RULER-v2 scores, plus lower AgentPPL and LongPPL Why build a new architecture? Agentic inference is a very different workload. Each round, a short action can return a long observation that needs to be prefilled, while the context keeps growing. That puts prefill cost, KV-cache size, and retrieval accuracy on the critical path at the same time. HySparse2 tackles all three with two levels of KV sharing: • KV Bridging: Following YOCO, full-attention layers in the cross-decoder build their K/V from self-decoder hidden states. • KV Reuse: Within each hybrid block, sparse layers reuse the preceding full-attention layer's KV cache and selection indices. Two more changes: token-level selection replaces block-level selection, and a forced window of recent tokens replaces the separate SWA branch, so local and global tokens share one KV cache. Since all cross-decoder KV caches now come from the self-decoder, prefill can stop once the self-decoder finishes. Paper: arxiv.org/pdf/2609.26368

MiMo-V2.6-Pro debuts as the top open weights model on the Artificial Analysis Intelligence Index (46). At $0.13 per Intelligence Index task, it lands on the Intelligence vs. Cost per Task Pareto frontier @Xiaomi has just released MiMo-V2.6-Pro, an open weights model with major advances in intelligence over its predecessor, MiMo-V2.5-Pro (Intelligence Index: 26). Despite the improvement, it retains the same attractive pricing at $0.435 per 1M input tokens (with a 99% cache-hit discount) and $0.87 per 1M output tokens. This makes MiMo-V2.6-Pro one of the most cost-efficient models to deploy. MiMo-V2.6-Pro is an MoE model with 1.02T total parameters and 42B active parameters. Stay tuned for additional analysis of the model. Check out MiMo-V2.6-Pro full benchmarking breakdown here: artificialanalysis.ai

Introducing Xiaomi MiMo-V2.6 — Pro & Flash. Frontier intelligence, all the modalities, built in public. 🔹 Two omnimodal models, advancing through scaled reinforcement learning 🔹 Pro performs on par with Claude Opus 5 and GPT-5.6 Sol across most agent benchmarks 🔹 Pro scores 46 on the Artificial Analysis Intelligence Index — the highest among open-source models 🔹 Stronger coding, computer use, 3D reasoning and creative capabilities 🔹 Open model weights, technical report, RL environments and training code Blog：mimo.xiaomi.com/mimo-v2-6

🚀 Xiaomi MiMo Desktop is now in invite-only beta. Invitees also get free, limited access to next-gen MiMo models during the beta. MiMo Desktop is a desktop agent — not another chat window. It takes your raw materials — Office files, images, video, audio, even zips — understands the goal, plans the path, and ships finished, editable output. What it can do: 🔹 Previews you can touch. Games, dashboards, demos load live in-session — click, break, and keep iterating. No local frontend setup required. 🔹 Edit by selection. Click or box-select any region — describe the change, and only that part updates. Every revision is versioned, so you can roll back anytime. 🔹 Smart dispatch. The system routes tasks to the right model and agents for you — quality when it matters, speed when it doesn’t. No model-menu guesswork. 🔹 Browser & computer use. It drives your browser to research, fill forms, and extract assets, then feeds findings straight into the task. Full computer use — screen, keyboard, mouse, cross-app work — with record & replay for repeatable flows. 🔹 Long tasks stay affordable. Partial regeneration + cache hit rates up to 99% in-session. Seats are limited. Apply here: mimo-ai.xiaomimimo.com/deskt…

Xiaomi makes its IFA debut with our largest-ever "Human × Car × Home" smart ecosystem showcase. A diverse portfolio spanning every aspect of life, seamlessly connected over 1.16 billion devices worldwide. More than 380 products are showcased at IFA, while bringing AI beyond the screen and into the physical world. And Xiaomi CyberOne is here. Hall 6.2, Booth 101. Come find us. 🤖