JonathanFly's favorites

1.		Talkie: a 13B vintage language model from 1930 (talkie-lm.com)
		773 points by jekude 26 days ago \| 331 comments
2.		TADA: Speech generation through text-acoustic synchronization (hume.ai)
		103 points by smusamashah 73 days ago \| 27 comments
3.		NanoGPT Slowrun: Language Modeling with Limited Data, Infinite Compute (qlabs.sh)
		185 points by sdpmas 80 days ago \| 35 comments
4.		Improving 15 LLMs at Coding in One Afternoon. Only the Harness Changed (can.ac)
		832 points by kachapopopow 3 months ago \| 295 comments
5.		Moltbook (moltbook.com)
		1652 points by teej 3 months ago \| 5 comments
6.		TimeCapsuleLLM: LLM trained only on data from 1800-1875 (github.com/haykgrigo3)
		737 points by admp 4 months ago \| 314 comments
7.		I'm making a game engine based on dynamic signed distance fields (SDFs) [video] (youtube.com)
		451 points by imagiro 4 months ago \| 78 comments
8.		How to code Claude Code in 200 lines of code (mihaileric.com)
		816 points by nutellalover 4 months ago \| 239 comments
9.		Welcome to Gas Town (steve-yegge.medium.com)
		354 points by gmays 4 months ago \| 224 comments
10.		Two kinds of vibe coding (davidbau.com)
		137 points by jxmorris12 5 months ago \| 105 comments
11.		Prompt caching for cheaper LLM tokens (ngrok.com)
		306 points by samwho 5 months ago \| 72 comments
12.		History LLMs: Models trained exclusively on pre-1913 texts (github.com/dgoettlich)
		897 points by iamwil 5 months ago \| 421 comments
13.		Neural audio codecs: how to get audio into LLMs (kyutai.org)
		428 points by karimf 7 months ago \| 119 comments
14.		Why do LLMs freak out over the seahorse emoji? (vgel.me)
		734 points by nyxt 7 months ago \| 416 comments
15.		Show HN: A little notebook for learning linear algebra with Python (little-book-of.github.io)
		190 points by tamnd 7 months ago \| 46 comments
16.		Qwen3-Omni: Native Omni AI model for text, image and video (github.com/qwenlm)
		571 points by meetpateltech 8 months ago \| 142 comments
17.		Almost anything you give sustained attention to will begin to loop on itself (henrikkarlsson.xyz)
		773 points by jger15 8 months ago \| 223 comments
18.		VibeVoice: A Frontier Open-Source Text-to-Speech Model (microsoft.github.io)
		448 points by lastdong 8 months ago \| 170 comments
19.		The decline of high-tech manufacturing in the United States (waldrn.com)
		124 points by giuliomagnifico 9 months ago \| 192 comments
20.		FFmpeg 8.0 adds Whisper support (ffmpeg.org)
		1033 points by rilawa 9 months ago \| 325 comments
21.		Training language models to be warm and empathetic makes them less reliable (arxiv.org)
		358 points by Cynddl 9 months ago \| 375 comments
22.		Diffusion language models are super data learners (jinjieni.notion.site)
		218 points by babelfish 9 months ago \| 16 comments
23.		Open music foundation models for full-song generation (map-yue.github.io)
		121 points by selvan 9 months ago \| 74 comments
24.		Show HN: KVoiceWalk – Voice cloning for Kokoro TTS using random walk algorithms (github.com/robviren)
		13 points by robviren on May 21, 2025
25.		The unreasonable effectiveness of an LLM agent loop with tool use (sketch.dev)
		447 points by crawshaw on May 15, 2025 \| 320 comments
26.		[dupe] Embeddings are underrated (2024) (technicalwriting.dev)
		484 points by jxmorris12 on May 12, 2025 \| 150 comments
27.		ACE-Step: A step towards music generation foundation model (github.com/ace-step)
		109 points by wertyk on May 6, 2025 \| 52 comments
28.		Dummy's Guide to Modern LLM Sampling (rentry.co)
		228 points by nkko on May 4, 2025 \| 37 comments
29.		Llasa: Llama-Based Speech Synthesis (llasatts.github.io)
		168 points by CalmStorm on May 1, 2025 \| 22 comments
30.		Ask HN: Share your AI prompt that stumps every model
		440 points by owendarko on April 24, 2025 \| 633 comments
		More