Puter

The world will ask you who you are, and if you don't know they will tell you.

@michael_chomsky @theo How would they add all the telemetry and all the dark patterns in an os repo?

it may sound insane, but @theo is almost definitely right to suggest that Anthropic should open-source Claude Code. The benefits are obvious: -it would be the most-starred repo of all time by a very wide margin, easily surpassing OpenClaw -they would get a bunch of free engineering from some of the smartest people in the world, and engineering velocity would skyrocket -a lot of the many small bugs plaguing CC would be eliminated -they would receive an unprecedented level of goodwill -CC SDK usage would likely 10x The downsides definitely aren't as pressing now that the source has been leaked. Regardless, 2025 was the year of mediocre harnesses. This will not be the case in 2026. I expect great things from Anthropic, OAI, Pi, Nous, OpenCode, and Slate Because once users taste a great harness, it's hard to return to something that isn't nearly as good.

@mamagnus00 @browser_use Sent in dm

@PuterOnX @browser_use What’s the email handle

Introducing: Free Tier for Browser Use Cloud 🚀 We’re giving all agents their own cloud browsers! > Unlimited browser hours > Free proxies > Persistent authentication Let your agents try for free ↓🔗

@mamagnus00 @browser_use Also it wont allow me to signup says cant sign up blocked by admin, i tried both sign in and up with both google and gh login

@PuterOnX @browser_use What do you like about it?

@mamagnus00 @browser_use Free tier with proxies and persistence allows for great social media automation experiences. Apps from meta and x for eg were strict about having residential ips which made working with them very difficult

@OfficialLoganK Lovely, what does byte for byte mean here?

Introducing Gemma 4, our series of open weight (Apache 2.0 licensed) models, which are byte for byte the most capable open models in the world! Gemma 4 is build to run on your hardware: phones, laptops, and desktops. Frontier intelligence with a 26B MOE and a 31B Dense model!

@ThePrimeagen Their ethos on non oss models and all of the company brainwashed into thinking closed source models are the way to go for safety is the actual concerning bit, shits crazy yo

I cannot stop thinking about Anthropic today for some reason 1. They claim that they are a company that prioritizes safety first and that they are creating a model responsibly 2. we learned from the code leak that anthropic employs deceptive techniques by calling fake tools to throw off distillers... Is this lying pattern built into claude or just the harness running claude? What else are they lying about? I am a bit more concerned now.

@himanshustwts Totally blind to their own hypocrisy, a classic

> be anthropic > the entire Claude Code codebase leaked just yesterday > sign an MOU to collaborate on "AI safety research" the very next day dude what a timeline

@icanvardar Lmao, big oof if true

how is claude code the worst harness in opus 4.6 benchmarks 💀

@LukeParkerDev That is true actually gemini is decent and underrated but it just has terrible ux like the worst, its so hard to work with it on any provider even on zen

everyone is talking about gpt 5.4 reviewing opus's work finding every issue. no one is talking about 5.4 reviewing gemini 3.1 pro's work always saying lgtm

@Zai_org I want to try but my login is on my laptop and I cant fill the registration form

Introducing GLM-5V-Turbo: Vision Coding Model - Native Multimodal Coding: Natively understands multimodal inputs including images, videos, design drafts, and document layouts. - Balanced Visual and Programming Capabilities: Achieves leading performance across core benchmarks for multimodal coding, tool use, and GUI Agents. - Deep Adaptation for Claude Code and Claw Scenarios: Works in deep synergy with Agents like Claude Code and OpenClaw. Try it now: chat.z.ai API: docs.z.ai/guides/vlm/glm… Coding Plan trial applications: docs.google.com/forms/d/e/1FAI…

@DeryaTR_ >It has been a very frustrating waste of time dealing with it, and it has consumed all of my goodwill toward Anthropic. had the same realisation in December/January, it was literally depressing to use any claude models and hindered my work more than I want to admit

Presumably Anthropic had fixed the Claude Code bug, so I wanted to give them the benefit of the doubt one more time & compare its rate limits with Codex. I used this simple prompt for planning and building: “Build a cute pixel RPG to play on the iPhone.” Both my Claude Max account and OpenAI Pro account started with fresh 5 hour resets. After several follow-up prompts, Codex completed a decent, simple, but fully playable RPG with exploration, crafting, collecting, fighting enemies & NPC interactivity features for the iPhone. I still had 93% of my 5 hour rate limit remaining with Codex using 5.4 xhigh & had used only 2% of my weekly limit. Claude Code, by contrast, still had not produced even a minimally viable playable version, with most of the above features missing, after already consuming 80% of the 5 hour limit. I asked for one more revision, but ran out of the 5 hr limit before it could even complete that. I then waited for a new reset and tried again to see whether it could finish the minimal game. It used up another 12% of the 5 hr limit and claimed it had added all the missing features. Yet the game was exactly the same, still unplayable, and not a single one of those missing elements had been added. Claude had also already used 8% of my weekly limit, still without delivering a product that Codex had already completed. Given that Anthropic also did not even bother to reset the limits from the token eating bug, I have now decided to cancel my Claude Max subscription. It has been a very frustrating waste of time dealing with it, and it has consumed all of my goodwill toward Anthropic.

@loosenedspirit @AlexPalcuie @GroundhogStrat Whats this from ? Really concerning if its anthropic although all their evil things would instantly make sense

@AlexPalcuie @GroundhogStrat “blameless for me but not for thee”

I repeat this to every new joiner at Anthropic but it's worth repeating in public -- we have a blameless culture and no single individual is at fault when bespoke complex systems break at scale

@FFmpeg April fools bois

FFmpeg is moving to Rust 🦀 Our use of C and Assembly in FFmpeg has been an unacceptable violation of safety. FFmpeg will be running 10x slower - but we're doing it for your safety. All your videos will appear green - safety first, working software later.

@badlogicgames Texhnically you can export gdoc as markdown

is there something like google docs, but for markdown? i need a cloud based collaborative markdown editor please.

Due to legal pressure, Hyprland will be adding an age verification system in the upcoming 0.55 release. On first launch, users will be asked a few questions about popular memes to estimate their generation, e.g. MLG, 67, Skibidi Rizz, or Kilroy was here.

@__morse @chronark @claudeai @opencode And the slop and shit engineering but yeah

@chronark @claudeai @opencode opencode is claude code without all the bugs

now that @claudeai code is open source, why should I use @opencode anymore?

@ZixuanLi_ He rugged the repo with some stupid reason and now its worthless

Good way to start exploration: zread.ai/instructkr/cla…

@Neitsab_FR @baanditeagle @thdxr Lmao thought the same thing

@baanditeagle @thdxr Received a lawyer mail ? Site is not open.

claude code source is 512K lines opencode is 118K we're getting LOC mogged

@nbaschez I feel the loop only has to be performed 3-5 times ? I havent had the loop organically go beyond 4 times

My biggest challenge with vibe coding / agentic engineering lately has been getting stuck in what I call a "plan doom loop" - have AI write a plan - review myself, seems good - have AI review plan, it always finds something - repeat It drains my time and energy to determine how important the "findings" really are Who has solved this