“MANUS AI: HYPE VS. REALITY 🔍 @peakji (co-founder of @ManusAI_HQ) confirmed rumors: âś… Built on Anthropic Claude Sonnet, not their own foundation model âś…Has access to 29 tools and uses @browser_use open-source for browser control âś…User communicates with executor agent and not https://x.com/_philschmid/status/1899046957860979178
“The popular AI agent “Manus” launched in China is automating about 50 tasks, and the scenario is too dystopian. It’s said to be more accurate than DeepSeek. It can simultaneously perform SNS analysis, financial transactions, research, purchasing, and more. https://x.com/thinking_panda/status/1897951585990590469
“I think China’s second DeepSeek moment is here. This AI agent called ‘Manus’ is going crazy viral in China right now. Probably only a matter of time until it hits the US. It’s like Deep Research + Operator + Claude Computer combined, and it’s REALLY good. https://x.com/rowancheung/status/1898093008601395380
“I tested Manus AI. It’s the closest thing I’ve experienced to a truly autonomous AI agent. I can’t wait till this thing can use desktop apps like Premiere and Photoshop. It low key feels like baby AGI. Here’s my 10 minute review: https://x.com/bilawalsidhu/status/1898945929970843842
“Google released Gemma 3 – their new Open-Source multimodal model family Here is everything you need to know in one thread: Gemma 3 27B currently ranks 9th in the LMSLOP arena beating models like o1-mini and o3-mini, DeepSeek-V3, Claude 3.7 Sonnet and Qwen2.5-Max In comparison https://x.com/scaling01/status/1899792217352331446
“Finally had a chance to try Manus. It’s a Claude wrapper, but a very clever one. Runs into the same issues as general agents, including getting stuck, but also capable of some good stuff. eg “get me the 10k for apple and visualize it in different ways to show me trends& details” https://x.com/emollick/status/1899148219335983207
“Have not played with Manus (the much hyped agent from China) yet but this seems credible The current frontier LLMs are very good & their abilities have not been fully explored even by the labs making them. Too much waiting for mysterious new LLMs, too little pushing what we have” / X https://x.com/emollick/status/1898930212521283848
“Anthropic is having a good start to the year: They grew from $1B ARR to $1.4B ARR in the first two months of 2025. Their models are also powering Manus, the latest AI sensation to blow up on X, which we also get into. w/ @JuroOsawa @nmasc_: https://x.com/steph_palazzolo/status/1899498723010662449
“New Anthropic research: Auditing Language Models for Hidden Objectives. We deliberately trained a model with a hidden misaligned objective and put researchers to the test: Could they figure out the objective without being told? https://x.com/AnthropicAI/status/1900217234825634236
“MCP is all the rage these days, and I read something along the lines of “MCP is better than traditional APIs because agents will dynamically adapt if you change the number of args in a tool, so versioning is less painful” đź‘€ I might be way off the mark but…this is still a huge” / X https://x.com/jerryjliu0/status/1899489337412378909
“A key to AI agents is an ability to self-correct. The world is full of odd errors and weirdness, and if AI can’t recognize and problem-solve when it hits a wall, errors compound & the agent is useless It doesn’t always, but here is Manus hitting problems & improvising solutions https://x.com/emollick/status/1899192116443509139
“The team behind Manus partnered with Alibaba’s Qwen to develop a Chinese version of its autonomous agent The collaboration will integrate Manus with Qwen’s open-source models and computing infra This comes just after Manus went viral last week! https://x.com/rowancheung/status/1899713389439377678
“Manus AI is less like a second DeepSeek moment and more like a Baby AGI moment. Everyone saying “it’s just a wrapper” is missing the point. Product and network effects are your only moat.” / X
https://x.com/bilawalsidhu/status/1899190239455072281
“Google DeepMind releases Gemma 3 family of open models!! 1B trained on 2T, 4B trained on 4T, 12B trained on 12T, and 27B params trained on 14T tokens vision input with 400M SigLIP, context length of 128k 27B ranks 9th on LMArena, outperforming o3-mini, DeepSeek V3, Claude 3.7 https://x.com/iScienceLuvr/status/1899729481176138122
“Wow – Google expected to DOUBLE in value after Anthropic reaches its future valuation of only $14T 🤯 https://x.com/nearcyan/status/1899624995413950910
“Cognitive Computations presents, Dolphin-MCP – an open source flexible MCP client. You can use Dolphin, ollama, Claude, any OpenAI-compatible endpoints. I made it as simple and flexible as possible – feel free to add more model providers, if you like, and make a pull request! https://x.com/cognitivecompai/status/1899736936039825705
“Built a stock tracking app using @Replit for execution and @grok 3 for code, research & conflict resolution. I started both projects at the exact same time in Replit and Claude 3.7 Sonnet; Claude 3.7 Sonnet couldn’t keep up. I was feeding the same information to both, and 3.7 https://x.com/samuel_riccardi/status/1895721439845589183
“New 42 min deep dive: Using Claude, Gemini & Grok to build a 3D city simulation, dynamic video annotations, smart shot lists, and even a AR HUD overlay. Chapters below: 00:00 LLM Superpowers You’re Not Using 00:26 Claude 3.7 Overview 02:38 3D City Vibe Coding for Creatives https://x.com/bilawalsidhu/status/1898020814576066585
“We’ve made several updates to the Anthropic API that help developers process more requests and reduce token usage with Claude 3.7 Sonnet. https://x.com/AnthropicAI/status/1900234837283197122
“I built a quiz game, SEO audit tool, and pricing calculator in under 30 minutes. Bolt DIY + Claude 3.7 is RIDICULOUSLY powerful. 5 key takeaways: 1. Setting up the free version of Bolt on GitHub 2. Using OpenRouter to access Claude 3.7 Sonnet API 3. Switching between https://x.com/JulianGoldieSEO/status/1896938696391409897
Token-saving updates on the Anthropic API \ Anthropic https://www.anthropic.com/news/token-saving-updates
“Introducing OlympicCoder: a series of open reasoning models that can solve olympiad-level programming problems 🧑‍💻 We find that OlympicCoder models outperform Claude 3.7 Sonnet, as well as others over 100x larger đź’Ş Together with the models, we are releasing: https://x.com/_lewtun/status/1899574591171035390
“Perplexity API now supports MCP. You can use it give any of your favorite AIs real time information, eg: Claude. https://x.com/AravSrinivas/status/1899850017546129445
“Just built a REAL-TIME eye-tracking laser blaster game with @MikeDKirby using Claude Sonnet 3.7 on @Replit! 🔥👀 No hands, just EYES. Wanna try it? 👇 https://x.com/MarcinAI81/status/1895348854364742056
“Vibe coding killed scripting. Well, scripting as we know it. AI tools can help you build “disposable apps”—purpose built for one-time use. Check out my latest video on how I performed a blog migration using Replit & Claude, saving myself a day of work. https://x.com/mattppal/status/1897663470830502190
“We are building the largest source of fully managed MCP servers, with complete auth support. You can now integrate apps to your @cursor_ai, @windsurf_ai, or @AnthropicAI’s Claude desktop without dealing with infrastructure, reliability, or authentication complexities. Simply https://x.com/KaranVaidya6/status/1896970570056786378
“I feel like between using Claude Code, Manus, and OpenAI Deep Research, you can start to see a near future where narrowly focused agents in expert hands make a big difference on work. Still far enough from full autonomy and error-free use that expert use will matter, for now.” / X https://x.com/emollick/status/1899188294132642059
“🔥The Open R1 team just dropped OlympicCoder and it’s wild: – 7B model outperforms Claude 3.7 Sonnet on IOI benchmark (yes, 7B!!) – 32B crushes all open-weight models tested, even those 100x larger 🤯 Open-sourcing the future of code reasoning! 🚀 https://x.com/fdaudens/status/1899568227312906734
“entity recognition using Anthropic citations! https://x.com/hwchase17/status/1899151312803246103
“The Perplexity API Model Context Protocol (MCP) is now available. We’ve built an MCP server for Sonar, giving AI assistants real-time, web search research capabilities. Powered by Perplexity, Claude can now search the web and deliver real-time and accurate insights on demand. https://x.com/perplexity_ai/status/1899849114583765356
“Windsurf Wave 4 is here! Included in this update: 🖼️ Previews ✏️ Cascade Auto-Linter ⚙️ MCP UI Improvements ➡️ Tab to Import ↩️ Suggested Actions 🫶 Claude 3.7 Improvements 🤝 Referrals 🖥️ Windows ARM Support … and more. https://x.com/windsurf_ai/status/1897378545799979238
“Text generation 3.2 is out. Gemma 3 support ! And a lot of improvements around tool calling agenting. And also lots of work towards many hardware/many backend !” / X https://x.com/narsilou/status/1899813420007919925
