📺 Xiaomi Is Livestreaming a Production RL Run — Burn Rate Included @XiaomiMiMo is training its MiMo-V2.6 in public — not a demo, a live post-training dashboard where every step's reward, loss and benchmark movement is watchable in real time. One day in, the meter reads roughly $1.2M spent and 60B tokens consumed. As Zhihu contributor Kitt在进化 puts it: this isn't a training site, it's a money-burning site — about ¥4,000 a minute. His bigger point: this is rare, valuable data. Almost nobody publishes what frontier-scale RL actually costs. 1️⃣ What the dashboard actually shows Every training step exposes the internals most labs keep private. His reading guide: 🔹 Loss family: training loss, entropy (are answers diversifying or collapsing into one mode), gradient norm, and train/inference KL divergence. 🔹 Reward: a mix of test-case scoring and rubric-based reward. 🔹 dynsam (dynamic sampling): avg@n performance over repeated tries, plus pass-rate buckets — including samples never solved and always solved. 🔹 DeepSWE v1.1 as the headline probe (mini-swe-agent, avg@3): Pro at 63.7, Flash at 60.7 at the time of writing. 2️⃣ The cost ledger, out in the open From the dashboard he extracts the numbers the industry usually guesses at: 🔹 MiMo-V2.6 Pro: roughly $36 per million tokens of training. 🔹 MiMo-V2.6 Flash: roughly $8 per million tokens. 🔹 His rule of thumb: equivalent inference runs 30-50x cheaper than these training figures. 3️⃣ What's actually being trained The sample distribution is also public: about 1,500 prompts per step, ~70% of them coding tasks. The remaining third is split across general, visual, cybersecurity and chat tasks — which makes this, in effect, an agentic-coding-centric RL run with side dishes. 4️⃣ Why it's worth watching He notes a university lab livestreamed a training run days earlier, but at nowhere near this scale or frontier relevance. For anyone who wants to learn how production RL behaves — rewards, entropy, dynamic sampling, benchmarks moving step by step — this is a rare open classroom. His only complaint, half-joking: a livestream this good deserves a comment section. 🔗 Full Reading: #Xiaomi# #MiMo# #ReinforcementLearning# #PostTraining# #LLM# #OpenScience# #AI#

We're livestreaming demos of last week's launch. Sept 2, 10am PT. * Agents and MCP. Talk to your robotics data in plain language. * Semantic Search. Describe a moment and find it in images and videos. * Comparison Mode. Line up multiple robot runs side by side * Bring Your Own Storage. Run Foxglove on data sitting in your cloud.

We are changing the prediction market and livestreaming industry. CRSHMARKET lets users trade on the outcome of any livestream in real time. We offer markets across esports, with IRL streams, live events, and more coming soon. We’ve already processed 500K+ trades, reached 43K+ users, and facilitated $10M+ in trading volume. This is just the beginning of what we’re building. New to CRSH? 👀

Bali touchdown 🌴💛 We’ll be livestreaming some of the key sessions from Binance Clubhouse on Binance Square, so there is no risk of FOMO. Looking forward to catching up with the community this week. See you there!

🇯🇵 While I don't approve of livestreaming, the Nigerians who are harassing random tourists and local Japanese are actually worse. This livestreamer was just sitting down with a camera and they suddenly tried to attack him.

This is a video I recorded while livestreaming at Swag Live 🔥💯 Tomorrow at 10pm ( GMT+7 ) I will comback 😍❤️ Please join me and have fun 🥳✔️ My swag link:

SouljaCoin 🪙 solana:BSdj5Bh5QgZut5zmKEbLwATLpiwUpix9U6dNL6pzpump Music 🎵 Gaming 🎮 Livestreaming 📺 Internet culture 🌐 Community 🤝 We building something for the Soulja ecosystem.

We are single-handedly changing the prediction market and livestreaming industries. We built the world’s first livestream prediction market with 60-second settlements, allowing you to trade on your favorite streamers in fast, real-time markets. Our goal from day one was to change an entire industry. We’re on track to do exactly that. Stay tuned. 👀