Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

I pay for Muse Code but didn't want to live in a terminal. On Windows it's WSL-only. So I built Helicon: an open-source desktop app for the actual Muse CLI. • uses your existing muse login, no API key • talks to muse serve over MSP, so Muse stays... show more

harjot.co

3,225 subscribers

51,594 views • 2 days ago •via X (Twitter)

Education Science & Technology

Anya Rossi• Live Now

Private livecam show

13 Comments

harjot.co2 days ago

· Free, MIT. Windows still needs WSL2 because Meta ships no Windows binary, Helicon runs natively on Windows and routes through it. How it works, with diagrams:

Tanguy2 days ago

@Muse wrapping the cli so Muse stays the agent. terminal only was blocking real use

harjot.co2 days ago

@Muse exactly. and reimplementing the loop is where you'd lose the subscription; the plan only bills through Muse's own harness, so a generic client silently puts you on pay-as-you-go. keeping Muse as the agent means the UI costs nothing extra. terminal-only was the only gap.

Tanguy2 days ago

@Muse yeah. keep Muse as the agent or you eat payg

Wesley | AI Builder2 days ago

@Muse The UI/harness boundary is compelling: keep the agent native while making sessions resumable. I’d surface the last tool action and a workspace health check so desktop state never hides terminal state. Is approvals or resume the bigger daily win?

Iulian Lita1 day ago

@Muse nice, but now the CLI works natively on windows, no more WSL needed.

harjot.co17 hours ago

@Muse Yep! Helicon 0.12.0 shipped yesterday with native support. With Muse for Windows installed, it runs muse.exe directly: Windows paths, PowerShell for ! commands, no WSL. WSL still works if you prefer it.

David Smith17 hours ago

@Muse Is there anything like this for Mac?

harjot.co17 hours ago

@Muse It works on Mac as well!

Praveen Juge1 day ago

@Muse The UI is high quality considering how quickly you put it together! Great job! What does “Ultra” do in terms of effort here? I don’t remember seeing that in the CLI

harjot.co17 hours ago

@Muse Thanks! Ultra is the top effort level in Muse's session protocol, one step past Max. The CLI's picker stops at Max, so on muse-spark it may act like Max for now. I'm checking what the model actually receives and will hide it if it's a duplicate.

Praveen Juge16 hours ago

@Muse Was using the app whole day! It's very good! Nothing broke so far!

harjot.co16 hours ago

@Muse Love hearing that, thanks for trying it out! Also a follow-up on Ultra: Muse sends it to the model as Max, so it was a duplicate. It is removed in 0.12.3, and Max is the top level now. If anything breaks, please let me know!

Related Videos

Grok Build for Windows. Your terminal just got a dev teammate. Plans, writes, runs, debugs. Not suggestions. Execution. Multi-agent system. Arena mode ranking outputs. This is real coding inside PowerShell. Set it up on Windows: * Open PowerShell - run as Administrator * Install it - paste the command irm | iex * Go to your project - cd into your repo * Run grok - launch the CLI * Sign in - one-time login with SuperGrok * Start prompting - build, fix, refactor * Let agents run - parallel work, less waiting * Review outputs - diffs, rankings, results Windows dev flow. No excuses. xAI Grok

Grok Build for Windows. Your terminal just got a dev teammate. Plans, writes, runs, debugs. Not suggestions. Execution. Multi-agent system. Arena mode ranking outputs. This is real coding inside PowerShell. Set it up on Windows: * Open PowerShell - run as Administrator * Install it - paste the command irm | iex * Go to your project - cd into your repo * Run grok - launch the CLI * Sign in - one-time login with SuperGrok * Start prompting - build, fix, refactor * Let agents run - parallel work, less waiting * Review outputs - diffs, rankings, results Windows dev flow. No excuses. xAI Grok

Mario Nawfal

98,671 views • 3 months ago

Introducing ZEPHY — every AI model. One window. Zero terminal. What you can do without ever leaving a single window: → Chat with cloud OR local/offline models, with smart routing that automatically picks the best one → Pull any model straight from Hugging Face uncensored too, auto-fit to your GPU → Write & run code in the built-in code editor → Chain steps into workflows on the node canvas → Generate media → Manage assets with an integrated crypto wallet Open source · Windows / macOS / Linux · no keys, no card. Site live -

Introducing ZEPHY — every AI model. One window. Zero terminal. What you can do without ever leaving a single window: → Chat with cloud OR local/offline models, with smart routing that automatically picks the best one → Pull any model straight from Hugging Face uncensored too, auto-fit to your GPU → Write & run code in the built-in code editor → Chain steps into workflows on the node canvas → Generate media → Manage assets with an integrated crypto wallet Open source · Windows / macOS / Linux · no keys, no card. Site live -

Pemersatu bangsa

20,064 views • 3 months ago

Open source is so back. Zuck just announced Meta is opening the weights for Muse Glimmer, with Muse Spark 1.2 coming soon But a year ago, everyone doubted Meta's position in the AI race In an interview I did with him, he admitted where they missed the mark: "Llama 4 was not on the trajectory that I thought it needed to be on. It was in many ways a big improvement over Llama 3, but we weren't trying to be a bit better than Llama 3. We're a frontier lab. It wants to be doing leading work." So he blew it up and rebuilt: "I felt like I'd learned enough about how to set up a lab that I wanted to reformulate the work we were doing." His conviction behind the rebuild: "AI is gonna be the most important technology in our lives, in our lifetimes. Building the capacity to build absolutely leading models is gonna be really critical for unlocking creativity."

Open source is so back. Zuck just announced Meta is opening the weights for Muse Glimmer, with Muse Spark 1.2 coming soon But a year ago, everyone doubted Meta's position in the AI race In an interview I did with him, he admitted where they missed the mark: "Llama 4 was not on the trajectory that I thought it needed to be on. It was in many ways a big improvement over Llama 3, but we weren't trying to be a bit better than Llama 3. We're a frontier lab. It wants to be doing leading work." So he blew it up and rebuilt: "I felt like I'd learned enough about how to set up a lab that I wanted to reformulate the work we were doing." His conviction behind the rebuild: "AI is gonna be the most important technology in our lives, in our lifetimes. Building the capacity to build absolutely leading models is gonna be really critical for unlocking creativity."

Rowan Cheung

53,329 views • 1 month ago

🥧 Introducing GooeyPi - A GUI for the Pi Family of Agents. Supports: Pi, Oh-My-Pi (OMP), and Prime Agent Very excited to share this with you all. Being into local models, I wanted to have a separate harness for my local models so I don't have to constantly switch configs. And I didn't want to go back to the TUI (nerds). But, I didn't want to miss out on all the features that I know and love. Turns out, I didn't have to. I just needed to build it myself. So I did. Features include: > Fully featured GUI experience > Agentic Browser > Realtime Voice Agent > Voice transcription (local or API) > Computer Use > Automation Desk > Agent to Agent messaging > Ask Question Tool > Git Control > Terminal > Pets > and more I tried to capture all of the most popular features I could in this first pass, but we're only just getting started. Available in BETA on MacOS, Linux, and Windows. I need testers! I have only tested on Mac and Linux. If you want to help me with this project, please reach out. Available:

🥧 Introducing GooeyPi - A GUI for the Pi Family of Agents. Supports: Pi, Oh-My-Pi (OMP), and Prime Agent Very excited to share this with you all. Being into local models, I wanted to have a separate harness for my local models so I don't have to constantly switch configs. And I didn't want to go back to the TUI (nerds). But, I didn't want to miss out on all the features that I know and love. Turns out, I didn't have to. I just needed to build it myself. So I did. Features include: > Fully featured GUI experience > Agentic Browser > Realtime Voice Agent > Voice transcription (local or API) > Computer Use > Automation Desk > Agent to Agent messaging > Ask Question Tool > Git Control > Terminal > Pets > and more I tried to capture all of the most popular features I could in this first pass, but we're only just getting started. Available in BETA on MacOS, Linux, and Windows. I need testers! I have only tested on Mac and Linux. If you want to help me with this project, please reach out. Available:

am.will

158,691 views • 1 month ago

Mark Zuckerberg (Mark Zuckerberg) on Muse, Meta's personal AI agent I recently sat down with Zuckerberg for episode two of the Sources podcast. We go deep on Muse, Meta's new (and, in my experience over the last few days, very capable) agent that comes with its own virtual machine and can do much more than a chatbot. Zuck, in fact, is confident that Muse will pay for itself by making people money. We also discuss his views on the data center backlash, his concerns about the concentration of power in AI, Meta's recent youth-safety settlement, privacy concerns about Meta's smart glasses, and more. Timestamps: 00:00 Zuckerberg’s Vision for AI 08:34 Data Centers, Jobs and Local Communities 15:03 Why Meta Is Building Personal Superintelligence 19:14 How Zuckerberg Uses Muse 25:00 Muse Pricing and Business Model 30:31 Privacy and Security for Personal Agents 41:06 Who Will Win the Personal AI Agent Market? 44:31 Rebuilding Meta’s AI Lab 54:09 AI Safety and Working With Government 01:01:51 Smart Glasses and Privacy 01:05:50 Teen Safety and Social Media Limits Thanks to the show's premier sponsors: Granola, Mercury, and Atlassian

Mark Zuckerberg (Mark Zuckerberg) on Muse, Meta's personal AI agent I recently sat down with Zuckerberg for episode two of the Sources podcast. We go deep on Muse, Meta's new (and, in my experience over the last few days, very capable) agent that comes with its own virtual machine and can do much more than a chatbot. Zuck, in fact, is confident that Muse will pay for itself by making people money. We also discuss his views on the data center backlash, his concerns about the concentration of power in AI, Meta's recent youth-safety settlement, privacy concerns about Meta's smart glasses, and more. Timestamps: 00:00 Zuckerberg’s Vision for AI 08:34 Data Centers, Jobs and Local Communities 15:03 Why Meta Is Building Personal Superintelligence 19:14 How Zuckerberg Uses Muse 25:00 Muse Pricing and Business Model 30:31 Privacy and Security for Personal Agents 41:06 Who Will Win the Personal AI Agent Market? 44:31 Rebuilding Meta’s AI Lab 54:09 AI Safety and Working With Government 01:01:51 Smart Glasses and Privacy 01:05:50 Teen Safety and Social Media Limits Thanks to the show's premier sponsors: Granola, Mercury, and Atlassian

Alex Heath

233,402 views • 9 days ago

gemini 3.7 flash vs deepseek v4 pro 0813 vs muse spark 1.2 – on voxel city dioramas three models each built three crossy road-style 3d scenes – a construction site, a nyc intersection, a river with a drawbridge – as single self-contained html files the setup: Nous Research's hermes agent cli on OpenRouter, three.js skills preloaded, identical prompts per scene tasks: 1. construction site – tower crane on a working lift loop, paver laying fresh road, roller compacting it behind 2. nyc crossing – four-way intersection with a traffic light state machine, queuing cars, pedestrians crossing on the walk signal 3. river drawbridge – double-leaf bascule that lifts for tall boats, cars queuing at the barriers, animated water every scene: Three.js r185, box geometry only, a locked 20-color palette, four camera presets, and a day/night mode with bloom. one file, no build step, no assets models: Google DeepMind gemini 3.7 flash, DeepSeek v4 pro 0813, AI at Meta muse spark 1.2 muse and gemini finished every scene in two to three minutes. deepseek took 15 to 41 minutes per scene - build time, all three scenes #1 gemini 3.7 flash – 6m 43s #2 muse spark 1.2 – 7m 20s #3 deepseek v4 pro – 91m 25s - total tokens #1 muse spark 1.2 – 440,279 #2 gemini 3.7 flash – 713,855 #3 deepseek v4 pro – 20,957,568 - total price #1 muse spark 1.2 – $0.53 #2 gemini 3.7 flash – $0.56 #3 deepseek v4 pro – $4.57 - agent calls across the three builds #1 muse spark 1.2 – 12 #2 gemini 3.7 flash – 18 #3 deepseek v4 pro – 143 observations: • muse won two of the three scenes on looks with the smallest files in the test – 887 to 1,042 lines against gemini's 1,934 to 2,377. cheapest, fastest to a good frame, and shortest turned out to be the same column • deepseek burned 20.96m tokens – 29x gemini, 48x muse – across 143 agent calls. prompt caching is the only reason that cost $4.57: the cache discount absorbed roughly $30 of resent context • gemini was the only model whose files needed zero fixes to render – and the only one whose night mode is cosmetic. the sky never darkens and one camera button does nothing. clean code for a scene it never looked at follow thehype. for 24/7 ai news, analysis and breakdowns

gemini 3.7 flash vs deepseek v4 pro 0813 vs muse spark 1.2 – on voxel city dioramas three models each built three crossy road-style 3d scenes – a construction site, a nyc intersection, a river with a drawbridge – as single self-contained html files the setup: Nous Research's hermes agent cli on OpenRouter, three.js skills preloaded, identical prompts per scene tasks: 1. construction site – tower crane on a working lift loop, paver laying fresh road, roller compacting it behind 2. nyc crossing – four-way intersection with a traffic light state machine, queuing cars, pedestrians crossing on the walk signal 3. river drawbridge – double-leaf bascule that lifts for tall boats, cars queuing at the barriers, animated water every scene: Three.js r185, box geometry only, a locked 20-color palette, four camera presets, and a day/night mode with bloom. one file, no build step, no assets models: Google DeepMind gemini 3.7 flash, DeepSeek v4 pro 0813, AI at Meta muse spark 1.2 muse and gemini finished every scene in two to three minutes. deepseek took 15 to 41 minutes per scene - build time, all three scenes #1 gemini 3.7 flash – 6m 43s #2 muse spark 1.2 – 7m 20s #3 deepseek v4 pro – 91m 25s - total tokens #1 muse spark 1.2 – 440,279 #2 gemini 3.7 flash – 713,855 #3 deepseek v4 pro – 20,957,568 - total price #1 muse spark 1.2 – $0.53 #2 gemini 3.7 flash – $0.56 #3 deepseek v4 pro – $4.57 - agent calls across the three builds #1 muse spark 1.2 – 12 #2 gemini 3.7 flash – 18 #3 deepseek v4 pro – 143 observations: • muse won two of the three scenes on looks with the smallest files in the test – 887 to 1,042 lines against gemini's 1,934 to 2,377. cheapest, fastest to a good frame, and shortest turned out to be the same column • deepseek burned 20.96m tokens – 29x gemini, 48x muse – across 143 agent calls. prompt caching is the only reason that cost $4.57: the cache discount absorbed roughly $30 of resent context • gemini was the only model whose files needed zero fixes to render – and the only one whose night mode is cosmetic. the sky never darkens and one camera button does nothing. clean code for a scene it never looked at follow thehype. for 24/7 ai news, analysis and breakdowns

thehype.

29,033 views • 1 month ago

BREAKING: Replit just launched mobile apps. Now you can build in Replit ⠕ and submit straight to the App Store for review. Describe your app → build it on Replit → test on your phone → publish to the App Store. I built a unicorn pet app for my niece: - Worked with Replit Agent to build it out - Scanned a QR code to download and test on my phone - One click to send to App Store Connect for review - A couple submissions later, it's live in the App Store! 🙌 I didn't write a single line of code! Just described what I wanted to Agent. Amazing! 😅 Go build something!! 👇

BREAKING: Replit just launched mobile apps. Now you can build in Replit ⠕ and submit straight to the App Store for review. Describe your app → build it on Replit → test on your phone → publish to the App Store. I built a unicorn pet app for my niece: - Worked with Replit Agent to build it out - Scanned a QR code to download and test on my phone - One click to send to App Store Connect for review - A couple submissions later, it's live in the App Store! 🙌 I didn't write a single line of code! Just described what I wanted to Agent. Amazing! 😅 Go build something!! 👇

Manny Bernabe

32,109 views • 8 months ago

meta muse spark 1.1 vs gpt 5.6 sol vs fable 5 vs grok 4.5 meta recently dropped muse spark 1.1 – a multimodal reasoning model from meta superintelligence labs built for agentic tasks. key facts: • 1m token context with active self-management – the model compacts its own history and keeps only the steps needed for later work • trained to orchestrate multi-agent systems: as main agent it plans and delegates to parallel subagents, as subagent it sticks to its job and knows when to escalate back • computer use trained to pick between scripting and clicking – writes automation when it's faster, clicks when it's simpler, batches actions per step • first public api from meta: the meta model api is now in preview • benchmarks: sweeps the agent column – mcp atlas 88.1 (opus 4.8: 82.2), jobbench 54.7 (opus: 48.4), humanity's last exam 62.1 (1st). loses coding – deepswe 1.1 53.3 vs gpt 5.5's 67.0, swe bench pro 61.5 vs opus's 69.2 our test – 3 prompts, single-file html, three.js, fully procedural, no assets: 1. norwegian house cantilevered over a fjord in a snowstorm – transmissive glass wall, fully modelled interior 2. beijing siheyuan courtyard house in dawn fog – instanced roof tiles, dougong brackets, glowing paper windows 3. new mexico adobe pueblo in an approaching dust storm – deep window reveals, windward grit accumulation we ran the test on AI/ML API platform results: - cost #1 muse spark 1.1 – $0.20 #2 grok 4.5 – $0.51 #3 gpt 5.6 sol – $1.93 #4 fable 5 – ~$5.20 - output tokens #1 muse spark 1.1 – 41,868 #2 gpt 5.6 sol – 49,139 #3 grok 4.5 – 64,954 #4 fable 5 – 81,849 - lines of code #1 muse spark 1.1 – 1,799 #2 gpt 5.6 sol – 2,377 #3 fable 5 – 3,088 #4 grok 4.5 – 4,216 observations: • muse spark is the cheapest of the four by a wide margin – 2.5x under grok, ~26x under fable per run. output quality tracks the price • only 7.4% of its output tokens are reasoning (3,104 of 41,868) – the model barely thinks before writing. economic, not pedantic: it commits to the first plan and ships it • the low loc is not compression, it's omission – all three prompts demanded instancing, muse spark delivered it in one muse spark's code quality – reviewed by fable 5: upsides: 1. all three files run 2. the adobe grit effect is legit – shader injection via onbeforecompile, windward faces detect storm direction through a normal-dot-wind term and darken procedurally 3. the fjord glass is real meshphysicalmaterial with transmission and ior, not a transparent quad 4. the siheyuan properly instances barrel tiles, dougong blocks and courtyard pavers downsides: 1. in the fjord file the strafe vector is negated – press a, you move right; press d, you move left. exactly the key mix-up we kept hitting with this model 2. all three files ship the model's self-doubt as comments: "// actually yaw orientation: need correct" sits above a direction vector that gets computed, abandoned and recomputed – dead vectors allocated every frame, 60 times a second 3. the siheyuan registers two separate keydown listeners, one containing an empty if-block 4. snow "accumulation" on the norway roof is a sine wobble on a scale value, not accumulation 5. "instanced snow" became 3,500 plain points. zero dispose calls anywhere pattern: minimal reasoning, minimal code, minimal price. it nails the flashy requirements – shaders, transmissive glass – and quietly drops the boring ones: instancing, controls, cleanup. you get a demo that mostly runs and a control scheme you can't trust follow thehype. for 24/7 ai news, analysis and breakdowns

meta muse spark 1.1 vs gpt 5.6 sol vs fable 5 vs grok 4.5 meta recently dropped muse spark 1.1 – a multimodal reasoning model from meta superintelligence labs built for agentic tasks. key facts: • 1m token context with active self-management – the model compacts its own history and keeps only the steps needed for later work • trained to orchestrate multi-agent systems: as main agent it plans and delegates to parallel subagents, as subagent it sticks to its job and knows when to escalate back • computer use trained to pick between scripting and clicking – writes automation when it's faster, clicks when it's simpler, batches actions per step • first public api from meta: the meta model api is now in preview • benchmarks: sweeps the agent column – mcp atlas 88.1 (opus 4.8: 82.2), jobbench 54.7 (opus: 48.4), humanity's last exam 62.1 (1st). loses coding – deepswe 1.1 53.3 vs gpt 5.5's 67.0, swe bench pro 61.5 vs opus's 69.2 our test – 3 prompts, single-file html, three.js, fully procedural, no assets: 1. norwegian house cantilevered over a fjord in a snowstorm – transmissive glass wall, fully modelled interior 2. beijing siheyuan courtyard house in dawn fog – instanced roof tiles, dougong brackets, glowing paper windows 3. new mexico adobe pueblo in an approaching dust storm – deep window reveals, windward grit accumulation we ran the test on AI/ML API platform results: - cost #1 muse spark 1.1 – $0.20 #2 grok 4.5 – $0.51 #3 gpt 5.6 sol – $1.93 #4 fable 5 – ~$5.20 - output tokens #1 muse spark 1.1 – 41,868 #2 gpt 5.6 sol – 49,139 #3 grok 4.5 – 64,954 #4 fable 5 – 81,849 - lines of code #1 muse spark 1.1 – 1,799 #2 gpt 5.6 sol – 2,377 #3 fable 5 – 3,088 #4 grok 4.5 – 4,216 observations: • muse spark is the cheapest of the four by a wide margin – 2.5x under grok, ~26x under fable per run. output quality tracks the price • only 7.4% of its output tokens are reasoning (3,104 of 41,868) – the model barely thinks before writing. economic, not pedantic: it commits to the first plan and ships it • the low loc is not compression, it's omission – all three prompts demanded instancing, muse spark delivered it in one muse spark's code quality – reviewed by fable 5: upsides: 1. all three files run 2. the adobe grit effect is legit – shader injection via onbeforecompile, windward faces detect storm direction through a normal-dot-wind term and darken procedurally 3. the fjord glass is real meshphysicalmaterial with transmission and ior, not a transparent quad 4. the siheyuan properly instances barrel tiles, dougong blocks and courtyard pavers downsides: 1. in the fjord file the strafe vector is negated – press a, you move right; press d, you move left. exactly the key mix-up we kept hitting with this model 2. all three files ship the model's self-doubt as comments: "// actually yaw orientation: need correct" sits above a direction vector that gets computed, abandoned and recomputed – dead vectors allocated every frame, 60 times a second 3. the siheyuan registers two separate keydown listeners, one containing an empty if-block 4. snow "accumulation" on the norway roof is a sine wobble on a scale value, not accumulation 5. "instanced snow" became 3,500 plain points. zero dispose calls anywhere pattern: minimal reasoning, minimal code, minimal price. it nails the flashy requirements – shaders, transmissive glass – and quietly drops the boring ones: instancing, controls, cleanup. you get a demo that mostly runs and a control scheme you can't trust follow thehype. for 24/7 ai news, analysis and breakdowns

thehype.

135,556 views • 2 months ago

Mark Zuckerberg (Mark Zuckerberg) just explained how the new Muse agent’s secured virtual machine works. Every Muse user can have a confidential cloud virtual machine containing highly private personal data, while the system is designed so even Meta itself cannot see that content. "you don't want someone to have to set up their own computer or VM. You just want to be able to provision it in the cloud, but you wanted to have the security and confidentiality that you'd have if you had the box sitting under your desk like in your house. " Zuckerberg says he does not know of another agent product with anything close to this setup. ---- From "Sources Podcast and Alex Heath YouTube channel, (full video link in comment)

Mark Zuckerberg (Mark Zuckerberg) just explained how the new Muse agent’s secured virtual machine works. Every Muse user can have a confidential cloud virtual machine containing highly private personal data, while the system is designed so even Meta itself cannot see that content. "you don't want someone to have to set up their own computer or VM. You just want to be able to provision it in the cloud, but you wanted to have the security and confidentiality that you'd have if you had the box sitting under your desk like in your house. " Zuckerberg says he does not know of another agent product with anything close to this setup. ---- From "Sources Podcast and Alex Heath YouTube channel, (full video link in comment)

Rohan Paul

159,163 views • 9 days ago

Start on desktop. Continue on your phone Hermes Bots Mobile App now keeps your conversations in sync with the Hermes Desktop App. Start a conversation with one of your agents at your desk. Grab your phone. Open the same bot. Continue exactly where you left off. Same agent Same conversation Same context No starting over. No copying prompts between devices Your Hermes agents are no longer tied to the computer sitting in front of you. This is exactly why I started building the mobile app

Start on desktop. Continue on your phone Hermes Bots Mobile App now keeps your conversations in sync with the Hermes Desktop App. Start a conversation with one of your agents at your desk. Grab your phone. Open the same bot. Continue exactly where you left off. Same agent Same conversation Same context No starting over. No copying prompts between devices Your Hermes agents are no longer tied to the computer sitting in front of you. This is exactly why I started building the mobile app

Luke The Dev

67,409 views • 11 days ago

On Omarchy for the last few days. Wanted a dock, so I built it with an agent. Wanted an infinite canvas, so I built it with an agent. Want a new file manager, being built now. Yes it's not super polished yet. But there is no going back to a locked down OS. You build your own apps. You build your own OS. PS: Posting from a MBPro because it's 2:43AM and I can't begin to learn Kdenlive now.

On Omarchy for the last few days. Wanted a dock, so I built it with an agent. Wanted an infinite canvas, so I built it with an agent. Want a new file manager, being built now. Yes it's not super polished yet. But there is no going back to a locked down OS. You build your own apps. You build your own OS. PS: Posting from a MBPro because it's 2:43AM and I can't begin to learn Kdenlive now.

Zsolt Kacso

103,014 views • 22 days ago

I built a mobile app to check Paddle revenue (because they don't have one): 👉 - Use your Paddle API key (read-only and scoped) - Live data with beautiful and useful graphs built with native Swift UI. - Multi-account supported, unified revenue metrics. - Data stay on device, no server (api requests are sent directly from your phone) - Home widgets - I made it free to download on App Store (once it's approved) - Buy the source code for $19 and customize it however you want (save 5hrs of prompting if you try to do it yourself). Some interesting facts about this side project: - I vibe coded with 100% claude code remotely on my Mac Mini (with my AI assistant setup) in less than 24 hours. - I have read 0 line of code in this project and never opened Xcode myself. - My AI assistant designed the app with GPT Image 2, built the app with Swift UI, test it on simulator (via screenshots), send the test build to TestFlight for me to test, and invited me to the app store connect account so I can test on my phone, then the AI submitted the app to App Store and currently waiting for approval. - For the website, I ask it to come up with a domain name, I bought it via manually and give it access via Cloudflare API, the AI design and create a static website with GitHub, test it with lighthouse CLI, deploy via GitHub pages, config the domain DNS, deploy the website. - Then I sign up an account with Polar payment, create an API key and ask the AI to setup a store, add payment, link with the account, and add the payment to the website. The entire process happened in the last 24 hours with me only talking to the AI via Telegram. This is such a fun side project not only to create an app that I wish exists, but also to push the limit of what I can use AI for, and so far I'm very impressed. I'll create so much more apps! It feels like I have unlocked a super power.

I built a mobile app to check Paddle revenue (because they don't have one): 👉 - Use your Paddle API key (read-only and scoped) - Live data with beautiful and useful graphs built with native Swift UI. - Multi-account supported, unified revenue metrics. - Data stay on device, no server (api requests are sent directly from your phone) - Home widgets - I made it free to download on App Store (once it's approved) - Buy the source code for $19 and customize it however you want (save 5hrs of prompting if you try to do it yourself). Some interesting facts about this side project: - I vibe coded with 100% claude code remotely on my Mac Mini (with my AI assistant setup) in less than 24 hours. - I have read 0 line of code in this project and never opened Xcode myself. - My AI assistant designed the app with GPT Image 2, built the app with Swift UI, test it on simulator (via screenshots), send the test build to TestFlight for me to test, and invited me to the app store connect account so I can test on my phone, then the AI submitted the app to App Store and currently waiting for approval. - For the website, I ask it to come up with a domain name, I bought it via manually and give it access via Cloudflare API, the AI design and create a static website with GitHub, test it with lighthouse CLI, deploy via GitHub pages, config the domain DNS, deploy the website. - Then I sign up an account with Polar payment, create an API key and ask the AI to setup a store, add payment, link with the account, and add the payment to the website. The entire process happened in the last 24 hours with me only talking to the AI via Telegram. This is such a fun side project not only to create an app that I wish exists, but also to push the limit of what I can use AI for, and so far I'm very impressed. I'll create so much more apps! It feels like I have unlocked a super power.

Tony Dinh

43,922 views • 4 months ago

Finally, a proper chat UI for Hermes Agent (open-source)! Hermes ships an official dashboard, but it's primarily built for management, and its chat is just a terminal piped into a browser tab. Hermes Web UI is an open-source chat-first alternative. It's self-hosted and points at your existing ~/.hermes state, so there's nothing new to configure. - It's a native web chat, not a terminal in a tab - Sessions group by date with a context ring - Kanban renders the agent's task board - Spaces manages your workspaces - Skills panel lists the full catalog - Tasks panel shows cron jobs - Insights show usage and activity - Memory shows MEMORY and SOUL files - Logs tails the agent, gateway, and error logs The whole setup runs 100% locally, binds to localhost by default, and you reach it over an SSH tunnel or Tailscale from your phone. I have shared the Hermes Web UI GitHub repo in the replies. Do note that it's a community project, not official, so expect occasional rough edges (concurrent profile runs are blocked for now). To dive deeper into Hermes Agent, my co-founder wrote a full masterclass about it, covering the learning loop, the memory tiers, self-evolving skills, GEPA, and running multiple isolated agents. Read it below.

Avi Chawla

78,099 views • 3 months ago

Hermes agent just left the terminal. 𝗛𝗲𝗿𝗺𝗲𝘀 𝗗𝗲𝘀𝗸𝘁𝗼𝗽 dropped yesterday. native app for macOS, Windows, and Linux. for months Hermes was the agent that learned your projects, wrote its own skills, and built a model of who you are. all of it buried in terminal logs. now it has a window. the important part is that it's not a wrapper. it runs the same agent core, the same sessions, memory, and skills as the CLI. you can start a task in the terminal and finish it in the app without anything resetting. the state is shared across every interface, not copied between them. what the GUI actually adds: → streaming chat that shows live tool calls and inline reasoning instead of a spinner → a preview rail that renders pages, code, and images right beside the conversation → an artifacts panel that collects every file the agent has ever produced → remote gateway mode, so you can point the app at a VPS and run the heavy work elsewhere → skills, cron, profiles, and gateways managed point-and-click instead of through YAML → voice mode, drag-drop files, and inline image generation remote gateway mode is the one worth slowing down on. the agent runs 24/7 on a $5 server while you control it from your laptop like a local app. other agent UIs are chatboxes with a logo. this one shows the autonomy instead of hiding it, so you watch the skills load, the tools fire, and the artifacts pile up as it works. it was teased in Jensen's GTC keynote. MIT licensed, local-first, no telemetry. if you already run Hermes, download it and everything is already there. your chats, memory, and skills carry straight over. i wrote a full masterclass on Hermes Agent that walks through the SOUL. md identity layer, the three-tier memory system, the self-evolving skills loop, and how to run three specialized agents 24/7. desktop is the interface that finally does all of it justice. the article is quoted below.

Akshay 🚀

51,540 views • 3 months ago

.Moxie Marlinspike marlinspike: "I think that LLMs are one of the first technologies to actively invite confession. And so the idea behind Confer is to do something similar to Signal of building something that allows people to avail themselves of the utility and awesomeness of LLMs, but to do it in a private way. It’s an end to end encrypted version of AI chat where you can interact with AI chat like you would with chatGPT. But all your conversations and conversation history are private and only, so only you have access to them. It’s using open source models.” Moxie was hired by Meta to integrate Confer’s privacy technology in their new AI products like Muse, a decade after Moxie worked with Meta to bring the Signal Protocol to WhatsApp. *Recorded in January, in the middle of the ocean, from a floating sauna Moxie built himself.

.Moxie Marlinspike marlinspike: "I think that LLMs are one of the first technologies to actively invite confession. And so the idea behind Confer is to do something similar to Signal of building something that allows people to avail themselves of the utility and awesomeness of LLMs, but to do it in a private way. It’s an end to end encrypted version of AI chat where you can interact with AI chat like you would with chatGPT. But all your conversations and conversation history are private and only, so only you have access to them. It’s using open source models.” Moxie was hired by Meta to integrate Confer’s privacy technology in their new AI products like Muse, a decade after Moxie worked with Meta to bring the Signal Protocol to WhatsApp. *Recorded in January, in the middle of the ocean, from a floating sauna Moxie built himself.

Sabrina Halper

21,618 views • 7 days ago

pov: you're running 15 loops across in glm 5.2 in opencode, codex, claude, amp or pi and have no idea which one needs you, bring em all to opensessions cmux, superset - cool ideas, but i got tired of blank screens, glitchy terminals, and macos-only apps i just wanted to know when my agents finish, fail, or need me so i built a tmux sidebar. now re-written in rust with worktree support runs inside your actual terminal. on any OS. - agents status' across all your projects at a glance - forwarded ports per project - live threads across amp, claude code, codex, opencode no electron. no blank panes. no new app. extremely fast and light. open source.

pov: you're running 15 loops across in glm 5.2 in opencode, codex, claude, amp or pi and have no idea which one needs you, bring em all to opensessions cmux, superset - cool ideas, but i got tired of blank screens, glitchy terminals, and macos-only apps i just wanted to know when my agents finish, fail, or need me so i built a tmux sidebar. now re-written in rust with worktree support runs inside your actual terminal. on any OS. - agents status' across all your projects at a glance - forwarded ports per project - live threads across amp, claude code, codex, opencode no electron. no blank panes. no new app. extremely fast and light. open source.

Palani — oss/acc

13,971 views • 2 months ago

Anthropic's in trouble, again! They spent years building what's now fully open-source. What made Claude feel different from a normal app is that the agent could act inside the interface instead of only talking in a chat box. For instance, Claude Artifacts let an agent render real UI, charts, dashboards, and interactive components that assemble live inside the response. Every major AI product tried to replicate it. But the problem was that unlike reasoning, planning, tool-calling, etc., none of it shipped natively with LangGraph, CrewAI, or Google ADK. So teams started building an owned version that required engineering the entire interface layer from scratch. Most teams, however, just settled for shipping the agent as a backend API in a chat box since rendering the UI is only one piece of it. To actually make it work, the interface layer also needed real-time streaming, state kept in sync between agent and UI, conversations that persist across sessions, and reconnection when a user refreshes mid-run. CopilotKit🪁 is now the only open-source framework that actually lets you build your own full-stack Claude-like apps. It decouples the agent from the interface, talking over AG-UI (an open protocol for agent-to-user communication). Being a standard protocol, the frontend never needs to know whether it is talking to a LangGraph or a CrewAI agent. You can change the backend anytime and the UI will never notice. In practice, CopilotKit's interface layer gives several pre-implemented React building blocks that wire the agent directly into the app, like: - generative UI, so the agent renders real components instead of text - chat windows, sidebars, and popups, or a fully headless setup - shared state, so the agent and app stay in sync - human-in-the-loop approvals, where the agent waits before acting - persistent threads that store the whole session, including the agent-user interactions and generated UI, not just text And because that full history is captured, those interactions can feed a self-learning layer that also improves the agent from real usage over time. The interface layer that Anthropic spent years engineering in-house is now literally available to any developer/team. CopilotKit is open-source with 30k+ GitHub stars, and AG-UI, the protocol underneath, is already supported across every major agent framework: LangGraph, CrewAI, Mastra, Google ADK, and more. CopilotKit GitHub repo → (don't forget to star it ⭐ ) If you want to go deeper, I found a detailed breakdown by Shubham Saboo recently on the three Generative UI patterns, with implementation. Read it below.

Anthropic's in trouble, again! They spent years building what's now fully open-source. What made Claude feel different from a normal app is that the agent could act inside the interface instead of only talking in a chat box. For instance, Claude Artifacts let an agent render real UI, charts, dashboards, and interactive components that assemble live inside the response. Every major AI product tried to replicate it. But the problem was that unlike reasoning, planning, tool-calling, etc., none of it shipped natively with LangGraph, CrewAI, or Google ADK. So teams started building an owned version that required engineering the entire interface layer from scratch. Most teams, however, just settled for shipping the agent as a backend API in a chat box since rendering the UI is only one piece of it. To actually make it work, the interface layer also needed real-time streaming, state kept in sync between agent and UI, conversations that persist across sessions, and reconnection when a user refreshes mid-run. CopilotKit🪁 is now the only open-source framework that actually lets you build your own full-stack Claude-like apps. It decouples the agent from the interface, talking over AG-UI (an open protocol for agent-to-user communication). Being a standard protocol, the frontend never needs to know whether it is talking to a LangGraph or a CrewAI agent. You can change the backend anytime and the UI will never notice. In practice, CopilotKit's interface layer gives several pre-implemented React building blocks that wire the agent directly into the app, like: - generative UI, so the agent renders real components instead of text - chat windows, sidebars, and popups, or a fully headless setup - shared state, so the agent and app stay in sync - human-in-the-loop approvals, where the agent waits before acting - persistent threads that store the whole session, including the agent-user interactions and generated UI, not just text And because that full history is captured, those interactions can feed a self-learning layer that also improves the agent from real usage over time. The interface layer that Anthropic spent years engineering in-house is now literally available to any developer/team. CopilotKit is open-source with 30k+ GitHub stars, and AG-UI, the protocol underneath, is already supported across every major agent framework: LangGraph, CrewAI, Mastra, Google ADK, and more. CopilotKit GitHub repo → (don't forget to star it ⭐ ) If you want to go deeper, I found a detailed breakdown by Shubham Saboo recently on the three Generative UI patterns, with implementation. Read it below.

Avi Chawla

460,618 views • 3 months ago

Introducing GrokTerm 🖤 A Rust terminal built for Grok Build. A real shell, multi-tab sessions, MCP-ready, and native two-way Grok Voice. A voice agent in the terminal, not a chatbot floating above it. Chapters: 00:00 Intro to GrokTerm 00:17 Testing two way voice conversations 00:30 Create Tetris 00:39 Moon landing page 00:56 Create calculator 01:35 Move TUI tabs to the right 02:01 Open calculator in browser 02:22 Design calculator via voice 04:30 Open moon in browser 05:15 Redo Tetris in HTML 05:35 Check calculator 06:04 Moon images 06:36 Light mode (macOS) 07:10 HTML Tetris live 08:21 New agent · light mode 08:49 Calculator effects 10:08 Close tabs 10:24 Credits Why this is different: • Custom TUI host that embeds Grok’s own TUI in a real PTY, not a browser, not a fake terminal. • Multi-tab shell + Grok sessions with one shared control plane. • Two-way Grok Voice to open tabs, delegate work, and iterate live. • CLI on macOS (Linux builds available, not fully tested yet). • Standalone notarized GrokTerm app for macOS. This is a WIP (work in progress) and I expect to add many useful features as I learn how to make and modify TUIs.

Introducing GrokTerm 🖤 A Rust terminal built for Grok Build. A real shell, multi-tab sessions, MCP-ready, and native two-way Grok Voice. A voice agent in the terminal, not a chatbot floating above it. Chapters: 00:00 Intro to GrokTerm 00:17 Testing two way voice conversations 00:30 Create Tetris 00:39 Moon landing page 00:56 Create calculator 01:35 Move TUI tabs to the right 02:01 Open calculator in browser 02:22 Design calculator via voice 04:30 Open moon in browser 05:15 Redo Tetris in HTML 05:35 Check calculator 06:04 Moon images 06:36 Light mode (macOS) 07:10 HTML Tetris live 08:21 New agent · light mode 08:49 Calculator effects 10:08 Close tabs 10:24 Credits Why this is different: • Custom TUI host that embeds Grok’s own TUI in a real PTY, not a browser, not a fake terminal. • Multi-tab shell + Grok sessions with one shared control plane. • Two-way Grok Voice to open tabs, delegate work, and iterate live. • CLI on macOS (Linux builds available, not fully tested yet). • Standalone notarized GrokTerm app for macOS. This is a WIP (work in progress) and I expect to add many useful features as I learn how to make and modify TUIs.

Dan

20,206 views • 1 month ago

Meta just launched its 1st image model after Mark Zuckerberg’s AI shake-up. Muse Image is Meta Superintelligence Labs' first image generator after Muse Spark. They said the model will power new editing features in Meta’s Instagram photo app and be added to its marketer tools for creating platform ads. Meta had relied on Midjourney and Black Forest Labs for generation inside Meta AI. Muse Image brings that layer in-house, so Meta controls quality, cost, and product timing. Consumers get free access through Meta AI, WhatsApp chats, and Instagram Stories. Power users need Meta One or another monthly plan when free limits run out. Users can start from prompts, add photos, annotate edits, or sketch changes directly. Meta says it can erase photobombers, create QR codes, and keep visual text readable. Advertisers get variants through Advantage+ creative, with edits, style swaps, and brand-matched versions. Meta says internal tests trail GPT Image 2 but beat Nano Banana 2 on editing. This is another move in Meta's effort in trying to convert AI infrastructure spending into revenue beyond social ads.

Rohan Paul

20,377 views • 2 months ago