<div style="font-family:'JetBrains Mono','Courier New',monospace;color:#e7e7e7;line-height:1.65"> <h1 style="text-align:center;color:#fff">《▓ Quantified API ▓》</h1> <p style="text-align:center"> <a href="https://github.com/Admany/Quantified-API" rel="nofollow"> <img src="https://i.imghippo.com/files/fC4894ILw.png" alt="Quantified API Banner"> </a> </p> <p>Quantified API, or QAPI, is a performance framework for mods that do a bit more than just slap random async onto everything and pray it doesn't implode xd</p> <p>Its goal is pretty simple: stop mods from frying the main thread and provide one shared runtime that can properly manage CPU work, parallel tasks, caching, dependency graphs, and GPU acceleration without turning the game into a deadlock simulator.</p> <div style="border:1px solid #8e24aa"> <p style="font-size:18px;color:#f0c6ff">《▒ Welcome to the New Generation ▒》</p> <p>With the release of V2, Quantified API has been upgraded by a <strong>LOT</strong>.</p> <p>The scheduler, CPU runtime, caching system, task graphs, GPU routing, compatibility layer, supported platforms, build system, and general API design have all been heavily improved or completely rebuilt.</p> <p>Quantified API now also has full official documentation, covering everything from basic installation to advanced compute workloads, GPU routing, caching, migration, troubleshooting, and examples :DDD</p> </div> <p style="text-align:center"> <a href="https://www.admany.dev" rel="nofollow"> <img src="https://i.imghippo.com/files/xVcg8245v.png" alt="Explore the new Quantified API documentation"> </a> </p> <div style="border:1px solid #6a1b9a"> <p style="font-size:18px;color:#e1bee7">《▒ What It Actually Does ▒》</p> <ul style="padding-left:20px;color:#e1bee7"> <li><strong>Proper async scheduling</strong> with worker lanes, workload routing, telemetry, and sane fallback behaviour</li> <li><strong>Parallel execution</strong> for workloads that can be safely divided across multiple CPU threads</li> <li><strong>Dependency-aware task graphs</strong> for workflows where jobs need to run in a specific order</li> <li><strong>Backend-aware compute routing</strong> across CPU, Vulkan, and OpenCL</li> <li><strong>Duplicate task suppression</strong> so repeated identical work can be merged instead of executed hundreds of times</li> <li><strong>Advanced caching</strong> with memory, disk, persistence, expiration, compression, and refresh support</li> <li><strong>Automatic mod registration</strong> through Fabric, Forge, and NeoForge metadata</li> <li><strong>Live diagnostics</strong> through runtime telemetry and the Quantified webpanel</li> </ul> </div> <div style="border:1px solid #5e35b1"> <p style="font-size:18px;color:#d1c4e9">《▒ Vulkan and OpenCL Acceleration ▒》</p> <p>Quantified API provides Vulkan compute acceleration that simply <strong>works</strong> without requiring users to manually configure a bunch of random settings first.</p> <p>Vulkan is the preferred GPU backend, while OpenCL remains available as a compatibility and fallback option.</p> <p>GPU acceleration isn't blindly used for every tiny task either. Quantified API considers the workload, runtime state, available hardware, and preferred backend before deciding where the work should run.</p> <ul style="padding-left:20px;color:#c5cae9"> <li><strong>Isolated Vulkan probing</strong> so broken drivers or loader issues don't instantly brick the game</li> <li><strong>Deferred initialization</strong> so GPU startup doesn't fight with Minecraft while the game is still loading</li> <li><strong>Cleaner runtime states</strong> so failed or half-initialized backends aren't treated as ready</li> <li><strong>Automatic fallback routing</strong> from Vulkan to OpenCL and finally to CPU when needed</li> <li><strong>Device selection and diagnostics</strong> through the Quantified webpanel</li> </ul> </div> <div style="border:1px solid #2e7d32"> <p style="font-size:18px;color:#c8e6c9">《▒ Faster Runtime ▒》</p> <p>The runtime underneath Quantified API was heavily redesigned to reduce the amount of overhead added around the work mods actually want to execute.</p> <ul style="padding-left:20px;color:#c8e6c9"> <li>Lower overhead for tiny and unique tasks</li> <li>Much stronger duplicate burst suppression</li> <li>Cheaper parallel workload orchestration</li> <li>Massively faster execution for smaller task graphs</li> <li>Better handling of thread-safe and non-thread-safe workloads</li> <li>Improved persistent and asynchronous cache behaviour</li> </ul> <p>Basically, it now spends a lot less time managing work and a lot more time actually doing it. Crazy idea, ik xd</p> </div> <div style="border:1px solid #c62828"> <p style="font-size:18px;color:#ffcdd2">《▒ Performance Improvements ▒》</p> <p>Internal runtime benchmarks against the older <code>v1.4.4</code> release showed major improvements across all tested workload types.</p> <ul style="padding-left:20px;color:#ffcdd2"> <li><st