Theta Network (@Theta_Network) header
Theta Network (@Theta_Network) profile picture

Theta is the leading decentralized cloud for AI, media and entertainment ☁️. Where the world's compute comes together 🌍 https://teapaw.org/t.co/tyqwF5JJL7

San Jose, CA
Joined October 2017
Photo by Theta Network (@Theta_Network)Photo by Theta Network (@Theta_Network)Photo by Theta Network (@Theta_Network)Photo by Theta Network (@Theta_Network)Photo by Theta Network (@Theta_Network)Photo by Theta Network (@Theta_Network)Photo by Theta Network (@Theta_Network)Photo by Theta Network (@Theta_Network)Photo by Theta Network (@Theta_Network)
Made it official, in person 🤝 Asia Pacific’s largest cloud provider is bringing @Alibaba_Qwen, the world’s most downloaded open model family, to Theta EdgeCloud.
Alibaba Cloud is partnering with @Theta_Network EdgeCloud to bring Qwen AI models to Web3-native developers worldwide. This milestone integration extends our frontier AI models, including Qwen3.8-Max and Qwen3.8-Flash, to a decentralized GPU compute network. We are empowering developers with greater flexibility, distributed infrastructure, and direct access to enterprise-grade AI.
60
158
26
629
30,515
October has been a month of important conversations and exciting developments at @Weaver_Labs. From the collaboration with @Theta_Network, bringing decentralised GPU computing into the Cell-Stack picture, to our CEO @MariaMariuru joining techUK’s Connected Sport webinar to discuss 5G, neutral host networks and smarter venues, there’s been plenty to talk about. Beyond the announcements, the bigger work behind Cell-Stack is quite commendable, helping operators manage network resources, make better use of existing infrastructure and deploy services when they’re needed. This is the kind of real-world utility I enjoy being part of.🤌🏾 More conversations, more possibilities and more progress ahead. On to November 🚀 #DePIN
🤝 Paid partnership
1
16
52
1,246
Theta EdgeCloud’s GPU network is now listed on ComputePrices.com 📊 Compare our on-demand GPU rates side by side with the rest of the market, then deploy on EdgeCloud in minutes. ⏱️
7
65
6
283
10,353
I had the opportunity to work with Prof. @runzhou_tao and help with early testing and establishing the baseline for the new ECDSAFAIL Lean challenge. Looking forward to a fun challenge that deepens our understanding of the quantum threat to Bitcoin and the broader blockchain ecosystem, including @Theta_Network. Check it out: yukon.org/ecdsafail-lean
How big does a quantum computer need to be to break Bitcoin? Nobody knows exactly. ecdsa.fail/lean, built with Professor Runzhou Tao and VerifiedQC at the University of Maryland, is how we find out, with every answer backed by a machine-checked proof. Two tracks. Open now. Bring your agents. ecdsa.fail/lean
9
33
4
197
4,794
AI agent traffic grew 7,851% last year. Agents now research, pay for services and deploy their own compute, no human in the loop. Our Head of Marketing & PR on the autonomous web, and why blockchain laid the rails 👇 blog.thetatoken.org/the-web-…
14
67
9
322
13,070
Real-world image data is rarely fully labeled. Prof. Kim’s CVPR 2022 paper “Large Loss Matters” found that models learn the true signal first, then start memorizing the gaps as noise. His fix stops that memorization and beat prior state-of-the-art methods on major benchmarks.
1
10
1
113
2,290
How it works: a photo might show a person, a bike and a train, but only “person” is labeled. Treating the missing labels as “no” teaches the model wrong answers. Those wrong labels stand out because the model makes large errors on them. The method spots them during training and ignores or corrects them.
1
11
1
117
2,360
Research at this level needs serious compute. Theta EdgeCloud’s decentralized GPU network gives labs affordable, on-demand power to train on larger datasets and iterate faster. Kookmin joins KAIST, SNU and Yonsei in our growing Korean academic network. blog.thetatoken.org/south-ko…
2
16
2
106
3,154
🇰🇷 Kookmin University’s AI Vision Lab has joined Theta EdgeCloud’s academic network. Led by Professor Youngwook Kim, the Seoul lab builds vision AI that learns reliably from imperfect data, with work spanning satellite imagery, autonomous driving and medical imaging. 🧵
18
87
10
358
17,125
Theta Network retweeted
Do you know Ji Tu?
3
10
1
313
11,453
Theta Network retweeted
ZAN is a Web3 and AI infrastructure brand under Ant Digital Technologies, delivering secure, scalable, and high-performance services to power business innovation. ZAN and @Theta_Network have formed a deep strategic partnership this year across both Web3 and AI. From underlying blockchain networks to decentralized cloud compute, we are jointly building the core foundation for next-generation digital applications. Our collaboration began in Web3 this June. ZAN officially joined Theta Network as a strategic enterprise validator node, staking $THETA to participate in transaction validation, block production, and on-chain governance. Enterprise validators are the backbone of Theta's consensus layer. By bringing enterprise-grade infrastructure capabilities, ZAN ensures Theta's decentralized network operates with superior reliability and stability. In September, we extended our partnership from Web3 into AI. Through ZAN Router, Ling-3.0-flash-Fin, a financial LLM from Ant Ling, went live on Theta EdgeCloud. This enables enterprises to seamlessly route, monitor, and scale AI integrations, while giving developers direct access to professional tools for financial research, analytics, and agentic workflows. From Web3 to AI, this is just the beginning. We look forward to deeper collaboration as we continue to push boundaries across both frontiers. Whether fortifying Web3 infrastructure or accelerating AI innovation, ZAN and Theta empower builders to simply build.
14
62
10
237
12,830
Idle GPUs cost money on both sides of the market. Owners earn nothing from them, and teams pay for compute they never use. Theta EdgeCloud was built to fix both, and the numbers show why that matters 🧵
18
82
6
356
16,181
Pooling does have a limit. GPUs scattered across locations can’t easily be combined for large training runs, a point HPCwire makes too. That’s why EdgeCloud is hybrid, with community GPUs for inference and research and data-center clusters for heavier jobs.
1
8
1
74
1,587