Millet

On September 29th, according to LateLatePost, Mi posted the promotion list internally on September 22, and Rolfley, 31-year-old, was promoted to level 22. According to those close to millimetres, grade 22 is the highest in the milligrade system, and promotions have since been mainly reflected in job changes; there have also been promotions of several heads of operations. On the same day, MiMo published two models, MiMo-V2.6-Pro and MiMo-V2.6-Flash. Of which, Pro for complex projects and long-range missions, F..

On September 22nd, Mi was officially released yesterday and opened with the Xiaomi MiMo-V2.6 series, which contains both Pro and Flash original whole-state models. MiMo-V2.6-Pro received 46 points in Artificial Analysis Inteligence Index, more than Kimi K3 and Qwen3.8 Max, but closed-source models such as Claude Table 5.1, GPT-6 Astra still..

On September 18th, the director of the Mimo's big model, Roosevelt, wrote yesterday, and since April, when MiMo-v2.5 was opened, Mi has been silent for almost six months, taking time to study one thing -- how far can intensive learning go? MiMo-V2.6 is in the middle of intensive learning. The team has expanded its three capabilities: to calculate the environment and tools (Multi-Multi-Purpose Imagerial Intensive Learning..

News from September 9, Mitech announced on September 8 that the Xiaomi Mimo Desktop desktop client was open for invitation. Users who apply for pass-through can use MiMo-X-Pro-Pro-Preview and MiMo-X-Flash-Preview for free during the test period. Mimo Desktop receives materials in various formats, breaks targets and calls tools, and finally delivers results such as PPT, Office files, web pages and lightweight games that can continue to be edited. It supports..

On September 5th, according to AGI, the latest large model of millet has entered the internal run-off phase and will be released in September as soon as possible. This information has not yet been officially confirmed by Mi. On the same day, OpenCode Go online anonymous model Omen Alpha was opened for testing by subscribers; it was reported to be linked to MiMo V3, but there is no publicly available evidence to confirm model identity. Notifications received by users of MiMo-V2.5-Pro-UltraSpeed showed that this version..

On the 22nd of August, on the 21st of August, the official Mi account number “Beta Edition Assistant” issued the Xiaomi Miclaw Closure Bulletin. According to the bulletin, in March of this year, rice opened the official Xiaomi Miclaw seal and received attention from many users. After nearly six months of closed testing, the team obtained many useful comments and suggestions on AI-type products. Now, it was decided to close the seal on September 21, 2026. The user 's Xiaoomi piclaw will be..

On June 17, Mi officially released yesterday the official version of the cloud light quantification product Xiaomi Mimo Claw, pre-positioning the MiMo-V2.5-Pro flagship model for depths appropriate to OpenClaw framework, the Native Matching MCP tool call protocol, equipped with a million-scale super-long context, and the primary low-threshold experience of Mimo model and AI smart body capabilities. Officially, based on the MTP 3-tier decoded structure, the full-link job insulation efficiency increased by about 3 in OpenClaw standard intelligent workflow testing..

News of 8 May, yesterday, Mi AI laboratory released and opened-source multi-language voice cloning TTS model OmniVoice, and the team built 580,000 hours of training data based on 50 open-source data sets covering 646 languages. (b) In 24 languages, speech similarities and understandings go beyond multiple commercial systems

On April 27th, according to blogger @foodshop researcher Will, on today's Mi Investor Day, the Vice-President of the Mi Group and Chairman of the Technical Committee Tsumi gave a speech on the theme “Agent-type remodeling of rice and human-carnage ecology”. The organizers summarized the core view as follows: Agent remodels the paradigms of millet and `human-carrying all-ecosystems' and spends more than $60 billion on AI over the next three years, a figure that is just below the bottom and higher; Mi has the full capacity of the IA age: the base, the data, the model, the frame and the ecological layer; the remodeling of one side..

On April 25, according to a news report, in recent days, the director of the MiGm team, Roosevelt, gave a series of specific judgments about the current pattern of large model competition, the AI intelligent body paradigm shift, and the AGI process during an interview with "Language is the World". Roosevelt expects that AGI will be achieved in two years. She estimated that the current progress was about 201 TP3T and that this year it is expected to move to 601 TP3T to 701 TP3T. According to Roosevelt, there are many companies in the country, including Kimi and MiMo, that have base models over 1T parameters, and the gap between China and the United States at the pre-training stage is "basically..

On April 23rd, Mi announced today that the Xiaomi MiMo-V2.5 series of models had officially opened public measurements. The MiMo-V2.5-Pro and MiMo-V2.5 models are about to open the world. 1AI was informed from the bulletin that the Xiaomi MiMo-V2.5 series consists of MiMo-V2.5, V2.5-Pro, V2.5-TTS Series, V2.5-ASR, with a stronger reasoning, a more stable Agent, a longer context, stronger command compliance and vague command understanding, and..

On April 23rd, the mime OS3 Beta version yesterday pushed the "crawfish" smart body, Xiaomei piclaw, to the millet 17 series, REDMI K90 Pro Max. According to Mi, Xiaomi Miclaw can make a mobile phone an AI tool that, after understanding your intentions and giving you a mandate, can call on one side of the application, ecological ability, and can select system-level tools for your command. It can know you, remember you, and judge your vague task, and help you to do it. It still holds..

On April 10th, Mi Mi today announced that Xiaomi Mimo has now access to the top Agent framework worldwide, Hermes Agent, and is limited to two weeks. As official Mi, Hermes Agent is one of the world ' s most active and open-source Agent frameworks with the ability to evolve and inter-chamber memory, to build experience from the mission, and to use it more and more to support platform-wide communications. MiMo-V2-Pro calls and depth Agent-specific optimization with 1M context-based, original tools, and is fully..

On March 20th, at the end of last night's launch of Mi Spring, the founder, chairman and CEO of Mi spoke about AI. He indicated that Mi would actively embrace the AI era and invest $60 billion in the AI field over the next three years. 1AI noted that in the early morning of yesterday, Mi Mi announced the launch of three large models - MiMo-V2-Pro & Omni & TTS. These models have landed in Xiaoomi Miclaw, Mimo Studio, Kanshan Office, Mi Browser, through..

On March 19, Mi suddenly announced the launch of three large models - MiMo-V2-Pro & Omni & TTS - this morning. At present, the models have landed in Xiaomi Miclaw, Mimo Studio, Kanshan Office, and Mi Browser, which can be accessed free of charge for a week through OpenClaw, OpenCode, KiloCode, Blackbox, Cline. Meanwhile, Mimo Studio's online

On March 19, in the early morning hours of this morning, Mimo V2 focused on three models: the flagship language base MiMo-V2-Pro, the full-mod Agent base MiMo-V2-Omni, and the voice synthesis model MiMo-V2-TTS. MiMo-V2-Pro: Total parameter mass exceeding 1T, activated parameter 42B, using the Hybrid Attention structure, with a mixed ratio of 5:1 in the previous generation to 7:1 and support for 1M 's hyper-long context..

On March 7th, Mi announced yesterday the launch of the mobile Agent product "Xiaomi Miclaw" and launched a small closed test. This is the first attempt in the country to bring OpenClaw applications to the end of a cell phone, based on the Mimo-based mega-model, the first to support the Mi17 series. The Xiaomi Miclaw focus validates the implementation capacity of the Large Model in the Mi "Homeland Ecology" system and explores the path of the Model from "dialogue capacity" to "system-level implementation capacity". With the tradition of running in the application sandbox..

ON DECEMBER 18, A SERIES OF NEW NEWS ITEMS, SUCH AS GROUP INCOME AND THE LATEST ADVANCES IN AI TECHNOLOGY, WERE RELEASED AT THE OFFICIAL MEETING OF THE MI 2025 CARAN ECO-PARTNERSHIP YESTERDAY. THE MI GROUP EARNED 34.4 BILLION YUAN IN THE FIRST THREE QUARTERS OF 2025, AN INCREASE OF 32.51 TP3T; ADJUSTED PROFITS OF 32.8 BILLION YUAN, AN INCREASE OF 73.51 TP3T; AND MILLET'S 7-YEAR RANKING IN THE WEALTH WORLD, A RISE OF 500 IN 2025

On December 17th, a new MiMo-V2-Flash model was officially released and opened in Mimo-V2-Flash, first looking at performance: a total of 30.9 billion MiMo-V2-Flash, an active parameter of 15 billion, using an expert hybrid structure (MoE), and also able to bend wrists with these front-source models DeepSeek-V3.2 and Kimi-K2. Remove the "open source" label, MiMo-V2-Flash The real killer It's a radical innovation in architecture that pulls the reasoning to 150 to..

On December 10th, Read Jia News, it was reported that Mi had completed the original version of the development of the product “Mi Chat”, a smart question and answer assistant. This post is part of our special coverage of the World Health Organization (WHO). Mi Group Partners / Managing Director, Cellular President and Managing Director Lu Wei Bing, who was the rice brand manager, revealed on December 4th that Mi had already started pressurizing input on AI in the previous quarter, and that although it is not yet possible to disclose much, Mi has made far more progress than expected in the AI Large Models and Applications. Mi-chan..

ON DECEMBER 4TH, THE MI GROUP PARTNER/PRESIDENT, PRESIDENT OF THE MOBILE PHONE DEPARTMENT, AND EXECUTIVE DIRECTOR OF THE MI BRAND LU WEI BING, RELEASED THE QUEUE OF ANSWERING THE NET QUEST IN A TREMOR YESTERDAY, PROVIDING ANSWERS TO QUESTIONS ABOUT ROOSEVELT'S PARTICIPATION IN THE MI AND THE FUTURE OF AI'S NEW STRATEGY. LUWAY ICE REVEALED THAT THE PRESSURED INPUT ON AI HAD BEGUN IN THE PREVIOUS FEW QUARTERS AND, ALTHOUGH IT WAS NOT YET POSSIBLE TO REVEAL MUCH, THE PROGRESS OF MI IN THE AI LARGE MODEL AND APPLICATIONS WAS FAR BEYOND EXPECTATIONS. MI THINKS THAT THE DEEP INTEGRATION OF AI WITH THE PHYSICAL WORLD IS THE NEXT STOP OF SMART TECHNOLOGY. LUWAY BING ALSO SAID MIFEI..

On November 14th, in the news that yesterday, Mi and Tents officially cooperated, the two sides announced that the "Major" platform under the cloud and the Mi Applied Shop would be connected to smart distribution. It was described that the developer, after completing the creation and commissioning of the AI smart body on the tatterer, could mount a single key to the millimetre application store without additional development. The first page of the millet application store is also synchronised with the "AI Smart Area" where users can search or browse their intelligence directly and use it as if they were downloading App. Last May, the twilight was set to a one-stop AI smart-body creation and distribution platform

ON SEPTEMBER 29TH, THE AGENCY OMDIA LOCAL TIME FORECAST ON THE 16TH OF THIS MONTH THAT THE AI GLASSES MARKET IS GROWING STRONGLY AND THAT, DRIVEN BY THE ACCELERATED ENTRY OF FIRST-HAND CONSUMER TECHNOLOGY PRODUCERS SUCH AS GOOGLE AND MI, THE VOLUME OF THIS PRODUCT WILL REACH 10 MILLION FOR THE FIRST TIME IN 2026, AN INCREASE OF 971 TP3T. OMDIA IS EXPECTED TO DELIVER 5.1 MILLION AI GLASSES THIS YEAR, AN INCREASE OF 1581 TP3T COMPARED TO 2024; AND A COMBINED ANNUAL GROWTH RATE FROM 2025 TO 2030 AS A WHOLE. CAGR)

On September 19, Mi announced today that Xiaomi-MiMo-Audio, the first original-to-end voice model of open source, had for the first time achieved a generalization of ICL-based samples in the field of voice. According to Mi, five years ago GPT-3 demonstrated for the first time the ability to acquire In-Context Learning (ICL, context learning) through a self-returning language model + large-scale unmarked data training, while in the area of voice, the existing large-scale model still relies heavily on large-scale tagging data and is difficult to adapt to new assignments to human intelligence. And Xia..