Exciting update: There’s a new leader in the Agent Arena among open models. MiMo V2.6 Pro by @XiaomiMiMo has landed the #1# spot with +3.8% net improvement! Earlier this week we evolved the Bash Recovery signal in Agent Arena (more info in thread). With this update, the new top 5 open models are: - #1# MiMo V2.6 Pro: +3.8% - #2# Deepseek V4.1 Flash (Max): +3.0% - #3# Hy4 preview: +2.9% - #4# MiMo V2.6 Flash: +2.5% - #5# GLM 5.2 (Max): +2.4% MiMo V2.6 Pro ($0.11 cost per task) and Flash ($0.05 cost per task) are the only two open models on the Agent Arena Pareto frontier! - Pro outperforms both #2# Deepseek V4.1 Flash while costing 15% less, and #3# Hy4 preview while costing 71% less. - Flash is 1.3 percentage points below Pro but is 55% cheaper. Congrats to the @XiaomiMiMo team on this contribution to the open ecosystem!

We're launching our leaderboard for FrontierCode 1.1, a private evaluation from Cognition that tests models on completing coding tasks in a way that would actually get merged by maintainers of real-world repositories @cognition announced FrontierCode in June 2026 and updated it to version 1.1 in July. Cognition built the dataset with open source maintainers who spent more than 40 hours per task. Each task requires the agent to make a change to a real open source repository, then checks both unit tests and qualitative tests for what maintainers would require to merge the change. We run FrontierCode 1.1 Main, the primary 100-task set of the benchmark's 150 tasks, using our Stirrup open source reference harness for apples-to-apples comparisons across models. All tasks are private to avoid contamination. Each attempt is graded with both blocking and non-blocking criteria. An attempt that fails any blocking check scores zero, and one that passes them all scores between 70% and 100%. Unlike other coding evaluations, FrontierCode has a more holistic view of success based on real usage. Models need to pass unit tests, but are also assessed on behaviors like making excessive changes and large diffs, modifying files outside the relevant scope, or adding unhelpful tests. We think this approach is valuable in the current context: models are now very capable at coding tasks, but their changes can still require substantial cleanup before a maintainer would merge them. Key takeaways: ➤ Leading models still fail more than 40% of the time: Claude Opus 5.5 (max with fallback) leads at a score of 52% after passing blockers in 58% of its attempts. Nine of the 100 tasks were not passed in any attempt by any of the models we evaluated for launch ➤ The top of the leaderboard is tightly packed: nine models from six labs score between 43% and 52%. MiMo-V2.6-Pro is the leading open weights model at 45%, followed by DeepSeek V4.1 Flash (max) and Kimi K3 (max) at 44% ➤ GPT-6.1 Sol (max) offers the strongest performance for its cost: At $0.81 per task it scores 49%, matching Claude Fable 5.1 (max with fallback). This means it achieves similar performance to leading Anthropic models, at less than a tenth of the cost per task Thank you to @silasalberti, @ybenpan, @cl571128, and the rest of the team at Cognition for building FrontierCode and collaborating with us to bring it to Artificial Analysis - we’re working to add FrontierCode to our coding agent leaderboard very soon.

🚀 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:

We’ve been impressed with GLM-5.2 and so are introducing a $9.99/month subscription to give you 2-5x discounted access to it and other open weight models like DeepSeek, Kimi, MiniMax, Mimo, Qwen. Use it on Cline CLI & IDE with $1.99 special promo if sign up via: npm i -g cline

🚀 MiMo Claw is LIVE Flagship AI Model + Kingsoft Office Integration 🤖 Unlock the Full Power of Agents • Powered by MiMo’s flagship model for reliable long-running tasks • 40–60% lower token consumption than comparable solutions, delivering better long-term value 📄 Productivity, All in One Place • Supports Word, Excel, PowerPoint, and PDF workflows • Generate professional documents, spreadsheets, and presentations with built-in templates 🎁 Expanded Free Benefits • Free to get started • Daily usage expanded from 1 hour to 4 hours • Fully cloud-based with zero deployment required 💰 Limited-time subscription: ¥14.9/month 👉 Try Now: 👉 Subscribe:

BREAKING: MiMo V2.5 Pro (Thinking) takes 3rd overall out of open weights models on Design Arena. MiMo V2.5 Pro (Thinking) places 8 positions higher than MiMo-V2.5 on the overall leaderboard, landing in the same performance band as Claude Sonnet 4.6 on frontend coding tasks. Huge congratulations to the @XiaomiMiMo team on these improvements!