Hackernews posts about an RTX 3090

RSS feed for an RTX 3090

01
newShow HN: Benchmarks of continual learning model trained for a month on a laptop
▲ 3volotat 7d 💬 1
02
newShow HN: Castellan – A real-time strategy game for agentsrts.emersoftware.cl
▲ 3emersoftware 5d 💬 4
03
newLLM from scratch, part 28 – training a base model from scratch on an RTX 3090www.gilesthomas.com
▲ 540gpjt 333d 💬 121
04
newWe got 207 tok/s with Qwen3.5-27B on an RTX 3090github.com
▲ 165GreenGames 194d 💬 52
05
newBenchmarking Qwen 3.6 35B MoE (3B active) on an RTX 3090www.gilesthomas.com
▲ 16gpjt 99d discuss
06
newExtending Raschka's GPT-2: an MoE trained from scratch on an RTX 3090www.gilesthomas.com
▲ 2ibobev 50d discuss
07
newExtending Raschka's GPT-2: an MoE trained from scratch on an RTX 3090www.gilesthomas.com
▲ 1gpjt 51d discuss
08
newBenchmarking Qwen 3.6 35B MoE (3B active) on an RTX 3090www.gilesthomas.com
▲ 1ibobev 96d discuss
09
newShow HN: OS Megakernel that match M5 Max Tok/w at 2x the Throughput on RTX 3090github.com
▲ 6GreenGames 206d 💬 1
10
newShow HN: ZigNet: How I Built an MCP Server for Zig in 1.5 Daysfulgidus.github.io
▲ 4Fulgidus 365d discuss
11
newShow HN: Accidentally built churn detection AI (I don't code)github.com
▲ 2willster787 443d 💬 2
12
newShow HN: Localvoxtral – Local real-time dictation on macOS with streaming STTgithub.com
▲ 1T0mSIlver 249d 💬 2
13
newShow HN: I over-engineered a home security camera that uses an LLM and talksgithub.com
▲ 1thecal 237d discuss
14
newStory of How Im Running an Unlimited $6/Month AI Provider on 4x RTX 3090s
▲ 8yolo-auto 140d 💬 3
15
newShow HN: Federation of robots collaboratively train an object manipulation modelgithub.com
▲ 7jafermarq 654d discuss
16
newShow HN: I built a 2nd-order PyTorch optimizer for LLMs that runs on 16GB GPUs
▲ 2dnosoz 185d 💬 4
17
newShow HN: Don't trust me bro: fixing GPT-OSS (3.49B tokens, 1k GPU hours, 1x3090)github.com
▲ 2iamskeole 60d discuss
18
newShow HN: GLM-Image Online – 16B AR+Diffusion model for accurate textglmimage.online
▲ 1hugh1st 290d discuss
19
newRunning Qwen3:30B MoE on an RTX 3070 laptop with Ollamablog.kekepower.com
▲ 18kekePower 516d 💬 1
20
newShow HN: Building a local FIX protocol triage agent on an RTX 3070domgalati.substack.com
▲ 2dgalati 190d discuss
21
newShow HN: I built a real-time OSINT dashboard pulling 15 live global feedsgithub.com
▲ 312vancecookcobxin 237d 💬 123
22
newShow HN: AgenticSeek – Self-hosted alternative to cloud-based AI toolsgithub.com
▲ 122Fosowl 553d 💬 18
23
newShow HN: Real-time 4D Julia set navigation via gamepadbanditcat.github.io
▲ 17BanditCat 354d discuss
24
newShow HN: OpenCode Senses, An insanely fast and highly accurate vision plugingithub.com
▲ 8itsmeadarsh 79d 💬 2
25
newAsk HN: Is the antigravity web site high CPU on your OS?
▲ 3newsoftheday 320d 💬 2
26
newI made an AI for my native language Assamese
▲ 2Kabyanil 719d 💬 1
27
newShow HN: I Am 15 and Built a Dual Backend MLP from Scratch Using CUDA C++github.com
▲ 1muchlakshay 465d 💬 2
28
newShow HN: Tracking book mentions across podcast episodes
▲ 1steyeomans 326d discuss
29
newShow HN: Model-agnostic cognitive architecture for LLMsgithub.com
▲ 6HimTortons 348d discuss
30
newShow HN: Local task classifier and dispatcher on RTX 3080github.com
▲ 26Shubham_Amb 268d 💬 2
31
newFather fixes dead RTX 3070 with a jerry-rigged capacitor from an old radiowww.tomshardware.com
▲ 2speckx 113d discuss
32
newTell HN: Llamacpp now supports unified system RAM offloading on Linux
▲ 6dabockster 221d discuss