正在加载视频...

视频加载失败

Smart grids. Waste reduction. Optimized routes. Remote ops. 5G helps your business cut carbon while boosting results. That's tech that's good for your bottom line—and the planet. 🌱 #SustainableTech

500,417 次观看 • 1 年前 •via X (Twitter)

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Microsoft made 100B parameter models run on a single CPU. bitnet.cpp: The official inference framework for 1-bit LLMs. The math behind 1-bit LLMs is what makes them revolutionary. Traditional LLMs use 16-bit floating point weights. Every parameter is a number like 0.0023847 or -1.4729. When you run inference, you multiply these floats together. Billions of times. That's why you need GPUs, they're optimized for floating point matrix multiplication. BitNet b1.58 uses ternary weights: {-1, 0, 1}. That's not a simplification. That's a fundamental change in the math. When your weights are only -1, 0, or 1: → Multiply by 1 = keep the value → Multiply by -1 = flip the sign → Multiply by 0 = skip entirely Matrix multiplication becomes addition and subtraction. No floating point operations. No GPU required. This is why bitnet.cpp achieves: → 2.37x to 6.17x speedup on x86 CPUs → 1.37x to 5.07x speedup on ARM CPUs → 71.9% to 82.2% energy reduction on x86 → 55.4% to 70.0% energy reduction on ARM The speedups scale with model size. Larger models see bigger gains because there are more operations to simplify. A 100B parameter model running at human reading speed (5-7 tokens/second) on a single CPU. That's not optimization. That's a different paradigm. Why 1.58 bits? Because log₂(3) ≈ 1.58. Three possible values = 1.58 bits of information per weight. The key insight: These models aren't quantized after training. They're trained from scratch with ternary weights. The model learns to work within the constraint. No precision loss. No quality tradeoff.

Tech with Mak

23,202 次观看 • 5 个月前

🌎VerAI is leading the charge for a greener AI future! Traditional AI training in massive data centers burns through energy, emitting CO2 equivalent to thousands of cars.🚗☁️☁️ But VerAI changes the game by using idle CPU/GPU power from contributors’ PCs 💻 Starting from a waitlist base of 30K contributors, we’re tapping into resources already in use, slashing the need for energy-hungry servers and cutting carbon emissions while you earn $VERAI. With this growing community, we’re building a sustainable ecosystem that proves innovation doesn’t have to cost the planet. ☘️🌱🌞 🛠️On the tech side, our platform on Base layer-2 dynamically detects unused compute power in real time, ensuring efficient allocation for AI tasks. Contributors can monitor their impact and earnings hourly, while developers access affordable resources to build transparent AI. It’s a win-win: your PC powers the future of AI, and you get rewarded all with minimal environmental impact.🌲 📊Here’s a mind-blowing fact: If 750 million PCs globally joined VerAI, we could save approximately 62.95 million metric tons of CO2 per year by reducing reliance on energy-intensive data centers for AI training. This calculated estimate assumes 80% of global AI workloads shift to VerAI’s distributed network, using idle resources that would otherwise go to waste. That’s like taking 13.5 million cars off the road annually! Join the waitlist and help us make AI sustainable! 👉 #GreenAI #Earncrypto #BlockchainInnovation

VerAi

13,027 次观看 • 1 年前

The math behind a Christmas light business is insane: • 140,000,000 homes in America • Average install: $800-$2,500+ • Seasonal service = Recurring every year • Low overhead (ladder, lights, insurance) • 60-70% profit margins once you dial in ops • Most competitors can't even hang lights straight Here's the breakdown: Let's say you charge $1,200 average per home. Install 3 homes per day (totally doable). → That's $3,600/day. Work 5 days a week during peak season (Nov-Dec). → That's $18,000/week. Run it for 8 weeks. That's $144,000 in gross revenue in 2 months. And people will call you NEXT year to do it again. This is what people miss about "boring" businesses. They're not exciting. They're not innovative. Nobody's going to write TechCrunch articles about your Christmas light company. But the math? The math is insane. That same 140M homes? They also need their lawns mowed. Their gutters cleaned. Their windows washed. Their driveways sealed. Their pools maintained. The opportunity is massive, and there's room for everyone. This is exactly why I started a home cleaning business. Same massive market. Same recurring revenue model. But I could run it 12 months a year instead of just 2. I started with $2,000 and scaled to $40k/month in 9 months. I don't touch a mop. I live wherever I want. I work when I want. If you want to learn more about how I did this and the opportunity in home services… DM me “Biz” & we’ll chat.

Mike Cleans

537,284 次观看 • 9 个月前