@ergyusuf3410459

QuestBridge addresses a real need from my existing Minecraft localisation work: adapt quest text for Turkish and German players without damaging colour codes, variables or book commands. This new standalone MIT project does not copy, bundle or relicense the pre-existing AGPL mod. The browser calls a Node.js server, which keeps game syntax immutable, asks an open model for natural-language fragments, validates the response and exposes a bounded cache. I ran Qwen3-4B-Instruct-2507 Q4_K_M through llama.cpp on my AMD Ryzen 9 7950X3D CPU. Four final unique live requests produced two format-valid drafts and two unchanged-English rejections; a repeated accepted draft hit the cache without another inference. Five focused tests passed. Raw evidence includes unsuccessful earlier trials and linguistic limitations. Turkish draft quality is weak and all output needs review; this is not a verified translation-quality claim. Token Factory cloud assignment is still pending, and no AMD GPU speed is claimed. The intended user is a modpack localiser reviewing reusable drafts, not a player trusting automatic unreviewed text. The measured benefit is avoiding a model call on repeated accepted input. AI coding assistance was used; code, original samples and evidence are public and reproducible.

Six Matches. One leaderboard. Build on AMD, earn points and compete for prizes.

🌎 Hybrid Hackathon | 💻 Online Build: 12-17 October 2026 | On-site phase: 17-18 October 2026 To participate in the hackathon, simply click Sign up with AMD. If you are not already a member of the AMD AI Developer Program (ADP), you will need to create an account. The AMD Developer Hackathon: ACT III is a hands-on event for developers, founders, engineers and builders who want to push what’s possible with AI on real infrastructure. 📍 If approved for the on-site experience, you'll also have the opportunity to showcase your project live. Travel and accommodation will not be covered.