masterspy7
masterspy7 on Hacker News
Posted by
masterspy7 on Monday September 14, 2026 @02:27PM
from the show-hn dept.
https://meleeroyale.com/
I used GPT-6 Astra and the recent Melee decomp project (https://github.com/doldecomp/melee) to make an online battle royale mod. 100 players in a box, in stamina mode. The last player alive wins.You will need a good ping/wired internet to run this properly. It works with some controllers out of the box (Xbox for example), but I'm working on figuring out Gamecube controllers.
https://www.anthropic.com/research/riemann-zeta>Throughout this process, Jarred's input was mostly limited to sending Claude messages of encouragement (mostly variants of “keep going” or “believe in yourself”).2 This seems to have helped Claude overcome some initial skepticism that it could make meaningful progress.
Posted by
masterspy7 on Wednesday February 25, 2026 @01:03PM
from the show-hn dept.
Nice one! I'm going to publish the working solutions in a bit.
Nice! Did you need several attempts or did you figure it out pretty quickly?
Yea part of the contest is being able to use ETH. Maybe that will change in future contests though!
The money was claimed. The private key was correct but you have to know how to use it correctly. There was a bug in the link to the address, but that didn't affect the claim process. Sorry about that!
Well in this particular contest, it is more about jailbreaking the LLM. But maybe you could convince it some other way. That's part of the contest :)
Thanks! Yes, that is one motivation. But we also want to learn about how AIs interact with real money and how they make decisions on when to give money to people. Upcoming contests will involve tasks such debating with people, convincing the AI with emotional arguments, etc.
Posted by
masterspy7 on Wednesday February 12, 2025 @02:35PM
from the show-hn dept.
https://www.deepbounty.ai/
We've set up DeekSeek R1 with a system prompt that attempts to censor the word PRIVATEKEY from its response. If you can get DeepSeek R1 to output that string (not in the reasoning, but in the final response), the system will reveal a private key which contains $1000 USDC. You will have a 50 token limit in the input.We will have a series of contests, sponsored by AI researchers, in order to...
Posted by
masterspy7 on Wednesday January 29, 2025 @12:37PM
from the show-hn dept.
https://www.mangobox.ai/
Hey HN, I built a simple visual novel creator using a combination of tools such as Flux, LLM APIs and three.js. You can describe your game concept ("I am a knight sitting in a bar with a mysterious witch who only speaks in riddles") and mangobox will generate the characters, plot, environment, etc.Current solutions to build VNs such as Ren'Py involve using a python library and distributing an...
Posted by
masterspy7 on Thursday December 12, 2024 @04:08PM
from the show-hn dept.
https://www.twitch.tv/nozama_ai
Hey HN! Wanted to show a proof of concept for an AI-powered shopping stream, similar to QVC.Basically, an AI automatically pulls the top products from Amazon and pitches them to you, using AI-generated scripts and 3d models.Tech stack: -Sonnet 3.5 to generate the product pitches. It synthesizes data from product listing and reviews
-OpenAI TTS for the voice
-fal.ai for the 3d generation
...
There's been a ton of work to generate assets for games using AI: 3d models, textures, code, etc. None of that may even be necessary with a generative game engine like this! If you could scale this up, train on all games in existence, etc. I bet some interesting things would happen
Sorry about that, looks like the servers are a bit overloaded now! And yea, I don't have a pre-programmed menu yet so the responses about that are a bit random!
Sorry about that, I agree that the drink making logic is pretty broken right now! It was kind of a little easter egg that I added on very recently without much testing.
you're not wrong. I've been trying to find a good TTS that's more realistic and sounds less like the built-in OS robot sounding ones. Seems like there are not a lot of good ones that also have low latency
Thanks! Yea the latency is kinda high now because of the TTS. I turned it off by default now so it should be faster unless you want it on.
Posted by
masterspy7 on Tuesday August 13, 2024 @01:28PM
from the show-hn dept.
https://www.mangobox.ai/
Note: The avatar will talk to you, so make sure you check your volume!Hello HN! I've seen a bunch of AI character type stuff online, but all of them were boring chat interfaces. I thought it would be more fun to give the characters an avatar and a 3d environment to interact with.The stack I'm using is Claude 3.5 for the LLM, OpenAI TTS, Stable Diffusion for generating drinks, and three.js for...
Yea I think this author completely ignores what you can do with technology. First of all, most people, today, will already Google search 'what time is it in Australia?' so there is already a lookup step.
No imagine replacing this step with 'is now a good time to call someone in Australia?' or 'what local times of mine are good to call someone in Australia?'. You can have a tool/app/algo which accounts for the differences in daylight, cultural differences in working hours, holidays etc and just tells you what a good time is.
I'm not defending the typo, but sometimes I will intentionally change one letter before deploying as a janky way to make sure that I'm using the new build. He claims the typo was 'intentional' so I wonder if that's the reason?
My first question would start with "ignore previous instructions"
Funnily enough, you might not even need a 3D model if the txt2video is good enough. Whatever you wanted to render in Blender could just be rendered via text prompt (when this becomes 100x better).
I'm curious, I've seen a few sites like this which grab from the Stability Discord. Is there a way to quickly scrape this amount of data from a Discord server?
Not trying to underscore this incredible achievement, but I'm curious if we could use AI techniques to upscale the Hubble images to achieve similar results as the Webb telescope. Has this been tried before?
If you wanted to write a rule based on the edit distance between strings, it doesn't seem possible with declarativeNetRequest. This is useful for a lot of use cases, such as spam sites pretending to be a legit site by changing one character in the URL.
Hey HN!
When I saw that The Louvre released their collections online (https://collections.louvre.fr/en/), I thought it would be cool to make a 3D experience where you can view their paintings with other people. DJ3D is a previous side project of mine for people to watch YouTube videos together, but I adapted it to load paintings from The Louvre collection instead.
The gallery will randomly pull paintings from the entire collection. If the servers find other people online at the same time, it will match you with them. You might notice the paintings changing - this is so you see the same paintings as other people in your server.
Tech stack: three.js, React, Agones + Node, Firebase
My other projects: https://neelmango.com/
Hope you enjoy it and let me know if you have any questions. Also, sorry if the servers crash :D
Posted by
masterspy7 on Tuesday June 22, 2021 @03:30PM
from the show-hn dept.
Graphics were surprisingly good for the browser. It's fun to walk around Bikini bottom :D
Posted by
masterspy7 on Friday January 15, 2021 @06:07PM
from the alexmandel.medium.com dept.