🍜客官，您的面来咯！ 🤖Naptha AI：让AI更聪明、更协作的新玩法 去中心化AI平台 Naptha AI 最近完成了600万美元的融资，这不仅让它成为 #AI 领域的一个关注焦点，也揭示了未来 AI 可能的发展方向。@NapthaAI 和传统 AI 不同，Naptha AI 的目标是让多个 AI 智能体组成一个庞大的网络，协作完成更复杂的任务。这种多智能体系统让 AI 不再是孤军奋战，而是像一个协作团队，为更丰富的应用场景提供可能。 🟡让多个智能体自由合作 Naptha AI 的核心优势在于它的多智能体系统。这种系统支持开发者创建数百甚至数千个可以互相沟通、协作的 AI 智能体。 例如不同 AI 代理人可以分别承担特定的工作，例如在医学研究中，一个代理人负责数据分析，另一个负责模式识别，而其他代理人则负责结果验证。这样一来，复杂的工作被细化分配，每个代理人负责自己擅长的任务，让效率更高，结果也更可靠。 🟡为什么选择去中心化？ Naptha AI 的另一个独特之处是它的去中心化架构。在这里不再依赖一个中央服务器来控制所有的智能体，而是让每个代理人都可以独立运行，彼此协作。这种方式不仅减少了单点故障的风险，还可以让数据处理更安全，适合需要高隐私的领域。这样的去中心化设计特别适用于 #Web3 和其他对安全性和独立性要求较高的应用场景。 🟡AI 和 Web3 的未来结合 Naptha AI 并不是在现有的大语言模型上堆叠更多数据，而是通过“多智能体合作”让 AI 系统更加智能、灵活。在未来，这样的 AI系统可以被应用在各行各业中，从科研到金融、从预测分析到医疗健康，不同的 AI 智能体分工合作，带来更多创新和效率提升的机会。 🟡团队的背景与愿景 Naptha AI 由 AI 和去中心化领域的两位资深人士 Mark Schmidt 和 Richard Blythman 创立。他们曾在各自的领域积累了丰富的经验，并以此为基础，将 Naptha AI 打造成一个支持多智能体高效协作的去中心化平台。用他们的话说，未来的 AI 不再是单打独斗，而是一个互相合作的智能生态，帮助人类解决各种复杂的问题。 🟡给开发者的机会 Naptha AI 不仅为AI技术的未来提供了一个创新的方向，也给开发者提供了广阔的舞台。现在，开发者可以在 Naptha 平台上设计和部署他们的AI智能体，在去中心化的环境中验证和调整这些智能体的协作效果。这对于想要探索 AI 新玩法的开发者来说无疑是一个极具吸引力的机会。

I tend to “put on different hats” to think about a single problem. Which to me feels similar to a single code base that makes calls to the same LLM with different prompts. Again, this is semantics so I can only share how I organize in my own mind, but for the sake of discussion around “multi-agent collaboration”, there’s tons more challenges in getting agents build by different people to collaborate, than different “agents” that share a code base and built by the same team. So in my mind, there is (1) simple agent (single mode), (2) complex agent (multiple modes), and (3) multi-agent (separate projects collaborating).

Hello @dAGIhouse Hackers, Users can now make queries to @NapthaAi agents directly from smart contracts, enabled through Oyster Serverless! 🦾 Docs: docs.marlin.org/user-guides/… Thanks to @richardblythman @roshanrags @PrateekReddyy Check out our progress on Sepolia:

To the 300+ PhD researchers at the frontiers of AI agents who just attended our event at NeurIPS- THANK YOU for making the night a huge success.

Some interesting ideas in this paper for extracting cultural memes from agent behaviour, and studying their propagation through agent societies. Anyone come across any other interesting approaches for meme extraction/generation? arxiv.org/abs/2411.00114

We are thrilled to have @richardblythman, the founder of @NapthaAI, sharing his expertise on decentralized computing for Multi-Agent Ops. 💫

I'll be giving a keynote about the Multi-Agent Approach to AGI at XYZ Paris on Feb 14th xyzparis.xyz/ May even get in one or two digs at Big Tech in preparation for the panels that follow... I have 10 codes for free tickets. Let me know if you'll be in Paris and would like one!

Today is a big moment. I'm very excited to announce our launch (along with fully open source release of the code) of v1 of the Naptha Stack. This is something we’ve been building towards for years. AGI won't be a monolith. We believe it will be a network of intelligent agents with heterogeneous models, architectures and data that adapt, compete and cooperate across many use cases. However, it's difficult to build such systems today because: 1. Agent and multi-agent frameworks built by different groups don’t interoperate. There are few standards for agent interactions. 2. Open Source Models use a variety of standards: Format for tool calling and reasoning, and adherence to structured output, vary depending on the model. 3. Scaling multi-agent orchestration across a network of nodes is an unsolved problem. The Naptha Stack solves this. Naptha is a framework and infrastructure for developing and running multi-agent systems at scale with heterogeneous models, architectures and data. Agents and other modules can run on separate devices, while still interacting over the network. This is a new type of platform that will enable new types of multi-agent applications, and we can't wait to see what AI engineers and researchers build. Come help us is to reimagine the internet. Let's bring forth the Web of Agents! Naptha Node: github.com/NapthaAI/naptha-n… Naptha SDK: github.com/NapthaAI/naptha-s… Docs: docs.naptha.ai/

Found some time to play around with @runwayml and made a first draft of a video for AI builders. Would appreciate feedback on how it comes across.

Agents Unleashed: Supercharging AI Agent Economies, the AI agent focused @ETHcc side-event during @EthCCweek looks like the best of the series so far, thanks in part to these Silver & Bronze sponsors! Here's more on them & the panels they'll appear on... lu.ma/osid96jj

In this week's MCP for Agents and Multi-agent Systems call we'll chat have the usual demos from the community (I'll be demoing our new automcp library for easily converting tools, agents and orchestrators from existing agent frameworks into MCP servers). We'll also chat about the implications of A2A for agents as MCP servers. 🥷 Find last week's recording and a registration link for this week below 👇

AI x Web3 is not just hype. Here are 20 use cases/ideas on what to build in this space right now. (Join our @dAGIHouse hackathon right before EthCC to build one of them, win a share of the $100K prize pool, and launch your product).

My prediction is that next year we see the first multi-agent simulation with 1 Billion agents.

We launched AutoMCP yesterday to easily deploy your existing AI agent codebases as MCP servers. Its like Vercel or Heroku for AI agents. We'd love to hear what you think about this product on ProductHunt, which went live today: 💬 Drop feedback 🛠 Tell us what you're building 🔺️ If you like it, upvote producthunt.com/posts/automc…

The year is 1610 and the American Frontier has just been opened. Do you build your new company/community (a) under the feudal system in Europe or (b) with sovereignty in the New World? The year is 2024. Do you build your AI project (a) under SB-1047 (and future AI regulation) in SF or (b) with sovereignty on decentralized infrastructure?

We know that AI x Crypto can raise capital, but it has failed to prove that it can use it to onboard AI developers. The reason is that financial gain is not the biggest driver for frontier AI developers. Read more in my piece in @Cointelegraph. cointelegraph.com/news/it-s-…

@NapthaAI helps developers build smarter, faster, and more collaborative AI apps. O0's job was to shape Naptha into a brand and help it stand out. Take a look at how we designed the brand with energy at its core. Explore the full case study here 👇 ozero.design/works/naptha-ai

OpenAI's Greg Brockman says the AGI future looks less like a monolith - and more like a menagerie of specialized agents. Models that call other models. “We're heading to a world where the economy is fundamentally powered by AI.” The goal is to unlock 10x more activity, output, and opportunity. AI lowers the barrier to entry. Domain expertise becomes leverage.

v1 of the @NapthaAI Stack has arrived! Interoperability, standardization, and distributed scaling for truly multi-agent systems Naptha founder @richardblythman joined @dongossen and @HoodAndLedger to break down whats coming. Lets go!

Switched from AutoGen and self-dev infra. to Naptha Stack - proper open source infra for multi-agent systems. Zero tokens, just distributed agent orchestration and interop. Been testing AutoGen but this feels way more robust for practical agent dev. 👨‍💻

🧵for interesting NeurIPS papers I come across this week. Also feel free to comment with any you've come across.

ML/LM Ops - the orchestration of models - is hugely underexplored in a decentralized setting. If the LM is a new computer, multi-LM Ops can facilitate the new internet. LM Ops => LM Co-Ops.

Excited to announce the @dAGIhouse online pre-conference and workshop series. Attendance is free! Go to this link to register: t.ly/dAGIonline Our program kicks off on June 11 and is designed to immerse you in decentralized AI, showcasing existing solutions and preparing you for the main event. Expect sessions with leading experts from companies like @valoryag, Flashbots, @CanonicalCrypto, and @NapthaAI, along with a workshop by @DoraHacks on succeeding at hackathons. Whether you're a Web3 or AI developer looking to explore dAGI, join us online to gear up for the hackathon, expand your knowledge, and connect with the community. 🗓June 11 – July 2 📍Online 🔗Register: t.ly/dAGIonline

We're running a real-life multiagent simulation! If you're at NeurIPS in Vancouver next week, come hang out with @NapthaAI, Plastic Labs @plasticlabs and @Nevermined_io to chat about the Web of Agents lu.ma/neurips24-mixer

Imagine a future where you can ask humanoid robots to clean your room, but some items, like heavy sofas, are too challenging for just one robot to move. Introducing CooHOI, a learning-based framework designed for the cooperative transportation of objects by multiple humanoid robots. 🤖🤼🤖 Our work has been accepted as Spotlight at NeurIPS 2024. Website: gao-jiawei.com/Research/CooH…

Olas Currents: Prediction Agents Join @david_enim & @tannedoaksprout (founding @autonolas members) THURSDAY at 4PM UTC with @richardblythman from @NapthaAI. AI agents are predicting the future, informing +450k txs on-chain. How accurate are they? nitter.app/i/spaces/1kvJpvWlOlLKE

🔥 Upcoming Naptha Workshops (online) 🔥 Learn why & how to build evolving multi-agent systems! 📅 Tuesdays ⌚ 6-7pm PT Our focus this week (Dec 3): 🗄️ USE CASES lu.ma/56ivm2gk - market simulations - social media coordination - open-source development + many more!

Unveiling the @autonolas Predict Leaderboard on @huggingface Created with @NapthaAI, this new benchmark is one of the first to evaluate the accuracy of AI prediction tools used by agents. It showcases tools which have up to 75% prediction accuracy! huggingface.co/spaces/valory…

Still guessing what to charge for your AI agent? We built a solution: The Nevermined AI Pricing Calculator Here’s how it works: 1️⃣Pick a use case – Chat, media generation, research… 2️⃣Customize components – models, tools, etc. 3️⃣Set your target margin ✅In seconds you’ll get: – Cost per interaction – Recommended Price – Revenue and profit projections Try it for free! 🔗: pages.nevermined.io/pricing-…

✨NEW EPISODE ✨ The AI crypto space is at a crucial inflection point after the recent Libra collapse. Don, Richard, and Robin discuss how the recent memecoin crash highlights the need to refocus on building real AI infrastructure rather than speculative tokens. The conversation reveals that while many AI crypto projects have faced significant drawdowns, there's substantial development happening beneath the surface - particularly in multi-agent systems and AI developer tools. We get into: – Why the recent Libra collapse triggered a broader AI crypto selloff – How early token launches can derail legitimate AI projects – The emergence of useful AI agent swarms and infrastructure – Why building for AI developers (not just crypto devs) is crucial – The future potential of agent-to-agent interactions – How crypto can enhance AI development and scaling – Why the intersection of AI and crypto remains promising despite market turbulence Tell us how we did: @dongossen @richardblythman @HoodAndLedger

From the fossils of the AI memecoins 🦕 We shall extract real naptha 🔥