Loading video...

Video Failed to Load

Go Home

Vibe Coding Deathmatch wednesday afternoon full video Bolt, Replit, Cursor, Windsurf head to head building the same project simple prompts across all, the results were: 🥉 Bolt + Windsurf 🥈Cursor 🏆 Replit Reasons: - Bolt was quickest, got into a 'no-audio' issue early and couldnt fix easily. first out,...

59,060 views • 1 year ago •via X (Twitter)

11 Comments

Ben Tossell's profile picture
Ben Tossell1 year ago

writeup will be on loom link: and most importantly, the link to play the DJ drums

TurboScribe's profile picture
TurboScribe1 year ago

What REAL customers say about TurboScribe's unlimited AI transcription 👇 🇺🇸 "Not only does it transcribe with amazing accuracy, it also filters out a ton of the unnecessary noise associated with pauses in audio. Keep up the great work!" - Kevin

Faisal Shariff's profile picture
Faisal Shariff1 year ago

It would have been good to see lovable as part of this too! Seems like they mostly all have the same issues. But I agree that I’m also most comfortable with Replit atm. Have you seen alarmingly that Tempolabs have added a ‘hire a dev feature’…what does that say about the confidence in the tech and the future of vibe coding 😥

Ben Tossell's profile picture
Ben Tossell1 year ago

Just a natural add on, people will get stuck and pay to get unstuck. Yeh I’m trying to switch up the tools periodically to see what’s good and not on others

Houman Shadab's profile picture
Houman Shadab1 year ago

Awesome stuff - reminiscent of the No Code Sports League I used to do (most Bubble devs)

Bryan Sanders's profile picture
Bryan Sanders1 year ago

Interesting matchup So who is Superfly Snooka?

Ben Tossell's profile picture
Ben Tossell1 year ago

? No idea

Jason Vu Nguyen's profile picture
Jason Vu Nguyen1 year ago

I agree with your death match results. If you’re not watching cursor carefully, it just goes off the rails.

Sam Jones's profile picture
Sam Jones1 year ago

Nice - this is helpful. Curious to know what transcription software are you using when recording voice instruction at the start?

Ben Tossell's profile picture
Ben Tossell1 year ago

And thanks!

Danny Quiroz's profile picture
Danny Quiroz1 year ago

@amasad Persistent redeploy button 🫡

Related Videos

I've built an AI agent that generates a logline of its choosing, creates text-to-video clips via 'scenes' it prompts with Luma, and then stitches them together automatically—even adding background music if you happen to have some present! At first, I let the model run wild—no direction, just "generate an interesting logline & prompts", basically. The results were quite interesting, but style was a bit all over the place (which was kinda cool in some ways actually). Then, I refined the prompt it was using to create the content—attempting to focus on cinematic feeling, character consistency (age, hair, ethnicity, name), and visual style. The idea is to build it out so that it can (hopefully) stick to genre, director influence, and a cohesive color palette. That will take some time, though. The agent is fully autonomous. It: Writes a story/logline and then the corresponding prompts Generates video clips Downloads them Stitches them into a final video (Optional) Adds background music all in one shot (it just takes ~5-10 minutes, and generates 17 clips or so at this time - I was using 30 as well but shortened it to test). You can upload your own background music to merge seamlessly, too. I'm working on more integrations to refine the output even further. Right now, I want everyone to try it out, but you'll likely need to upload your own API keys (because I'm broke 😂). Still figuring out the best way to share it! I'll have to explore Replit or Gradio, but that will be a little bit either tonight or tomorrow. The final video below shows where it's at as of right now. And in the comments, I’ve included different outputs—some completely chaotic, others more polished after prompt tuning.

Kiri

69,476 views • 1 year ago

I just compared Claude Code vs Codex vs Cursor CLI The task was to build a Next.js app with Tailwind 4 and shadcn components to collect customer feedback and showcase it with a widget. I gave all three the same prompt and let them go for 30 minutes to see what they came up with. Claude Code with Opus 4.1 Even though I told it to set up the app in the existing project folder, it tried to create a directory for it. After I interrupted and told it not to do that, it built a demo form and landing page with no errors. I had to ask it to make the demo interactive so users could submit a testimonial and preview it. The landing page looked like AI and was pretty basic, but it worked and it was done in a fraction of the time of the others. Total tokens used: 33k Codex with GPT-5 At the end of the 30 minutes I just could not get Codex to produce a working app. It got stuck in a loop of not being able to set up Tailwind 4 and despite many, MANY, attempts, I ended up with a "failed to compile" error. Total tokens used: 102k Cursor Agent with GPT-5 This was the slowest agent by far and a couple of times I actually thought it got stuck in a loop and was close to Ctrl+C'ing to cancel it. The TUI is really nice though, especially how it shows diffs and it did eventually build a working app (after one or two slight errors that needed fixing) The demo was interactive and it had a very minimal design that looked bare but also a lot less like an "AI generated" app than the Opus 4.1 design. It also wasn't too chatty and just did what it needed to do! Code quality was on a par with Opus 4.1, but it did use 5.5x as many tokens to get there. Still cheaper than Opus on a direct comparison but not when you factor in a Claude Code Max subscription. Total tokens: 188k I'll be able to do a proper comparison and record some videos when I'm back from holiday but for now, Opus is still the more capable model out of the box and Claude Code is the more complete CLI product. It will be interesting to see how Cursor evolve their CLI though with commands and subagents because I think with GPT-5 they have a real shot at providing competition for Claude Code if they can optimise output to get similar quality with less tokens. Jump to 0:40 in the video to see the two apps. Which do you think is which? ;)

Ian Nuttall

195,173 views • 1 year ago