Alternatives
Products that do what Profile does
Physics grounded, & cost aware, optimization loop for vLLM
- 1
- 2
- 3
- 4
- 5WF
2021 · prodfiler.com
- 6PI
2011 · profile.io
- 7IM
It's a chrome extension that automatically loads the specs from the Hugging Face model card into the calculation. > To test it, install the extension (no registration/key needed) and navigate to a HF model page. Then click the "VRAM" icon on the top right to open the sidepanel. You can specify quantization, batch size, sequence length, etc. Works for inference & fine-tuning. If it does not fit on the specified GPUs, it gives you an advise on how to still run it (e.g. lowering precision). It is inspired at my work, where we were constantly exporting metrics from HF to estimate required…
2025 · chromewebstore.google.com
- 8
- 9PS
2015 · github.com
- 10JM
2013 · github.com
- 11AP
2013 · github.com
- 12AW
I’ve spent the last few days prototyping a browser-based game engine stack designed for instant-load, "Roblox-for-Teens" style deployment. The goal is to lower the floor for AA indie devs who want high-fidelity physics and instant web distribution without the overhead of Unity or a heavy Godot export. The Stack Engine: A headless WASM/WebGPU implementation. I’m currently using custom flags and selective streaming to bypass standard browser memory and execution limits. Physics: I’ve ported Jolt (Guerrilla Games) but modified the solver to mimic the original Blam! engine (Bungie/Halo…
Jan 2026
- 13AS
Sep 2025 · github.com
- 14VV
2021 · vktracer.com
- 15IP
See https://github.com/rodgarrison/rdpmc If you've been interested in micro-benchmarking code using Intel's PMU HW capability to capture per HW core metrics, this repository may be for you. The repository includes 'doc/pmu.md' with a copy of Intel's software reference manuals (PDF) so, all in one place, you can understand and develop PMU profiling for your Intel micro architecture. Example included. Source code is 451 LOC with comments. Features: * Minimum, complete * Does not require yum install msr-tools * Header only: include and you're done * Works in user-space…
2022
- 16CW
Hi HN, I built ChartGPU because existing webgpu libraries are paid-for libraries like Scichart. F that. Keep it, I'll just build my own and make it open source :) The problem: I needed to visualize 1M+ data points for [your use case]. ECharts, Chart.js, and others dropped to single-digit FPS. The solution: Built a charting library from scratch using WebGPU. Key features: - Line, area, bar, scatter, pie charts - LTTB downsampling on GPU - Real-time streaming support - ECharts-style declarative API - React wrapper included Live demo: https://chartgpu.github.io/ChartGPU/…
Jan 2026 · github.com
- 17VV
2015 · github.com
- 18LH
I work on inference scheduling — KV cache-aware routing, load balancing across GPU workers, that kind of thing. I wanted something like k9s but for my inference stack. Nothing existed, so I built it. llmtop is a real-time terminal dashboard for LLM inference workers. It scrapes the Prometheus /metrics endpoints that vLLM, SGLang, and LMCache already expose and shows everything in one view: KV cache usage, queue depth, TTFT/ITL latencies (P50/P99 from histogram buckets), token throughput, prefix cache hit rates. Color-coded — red means go fix it. ``` brew install…
Mar 2026 · github.com
- 19WP
2023 · webgl-particle-life.netlify.app
- 20SO
A simple calculator that estimates how many concurrent requests your GPU can handle for a given LLM, with shareable results.
2025 · selfhostllm.org
- 21GM
Hey HN! This idea started with me not being able to buy a GPU and constantly losing to bots/scalpers. I figured I'd use this as way to see how far I can get with 'vibe-coding and designing'*. The end result was pretty far! Here are more details of behind the scenes. In a future blog post, I'll detail behind the scenes process of building this. - The landing page is React/Typescript/Tailwind.css (which I've never used before) - The dashboard is based on Evidence.dev - which is SQL queries in Markdown + little bit of custom Javascript for chart formatting (again never used…
2025 · gpuisfine.singhkays.com
- 22TR
Hi HN, Today I'm showcasing Trunchbull, a benchmarking platform designed for authoring benchmarks and running them against different models. We have direct support for benchmarks that use the harbor authoring system, custom tool authoring via the vercel ai sdk and configuration limits. We've also already imported terminalbench 2.0, as a sort of proof of concept that our harbor task orchestrator works, although you currently need a paid account as we are provisioning sandbox environments. I've made several popular benchmarks publicly available for testing. You dont need an account or your…
24d ago · trunchbull.dev
- 23CS
It's extremely difficult for founders, recruiters and hiring managers to screen their candidates for AI proficiency at scale. That's why we built Corepoints. You can create and send OAs where AI usage (with AI chat) is a core metric. You have full control of the testing environment: hallucinations, data leakages, LLM behavior + Grade candidates on aspects such as their answer accuracy (of course), prompting quality, reasoning quality, hallucination susceptibility, token usage, and more. We're currently doing a demo/beta run for about the next month or so that we can iterate off…
Mar 2026 · corepoints.ai
- 24CL
Hey HN, I wrote a compute shader emulator that started as a 50-line script to help me understand reduction shaders. What makes this implementation interesting is its technical approach; I've leveraged Nim's macros and closure iterators to simulate lockstep execution of logical threads. The emulator runs GPU compute shaders on CPU, simulating workgroups and subgroups with proper synchronization. It supports GLSL subgroup operations and provides nice debugging messages. The emulator works with Nim code that follows compute shader patterns I've documented some technical aspects here:…
2025 · github.com
Ranked by how close each launch is in meaning, then by votes. Refine with a description →