Hackernews posts about GPT-4 LLM

1. 01
   

   newAsk HN: How are LLMs getting smarter beyond just scaling?

   ▲ 2chenxi9649679ddiscusssavesaved
2. 02
   

   newChinese researchers build military AI using Meta's open-source Llama modelwww.tomshardware.com

   ▲ 5AnhTho_FR699ddiscusssavesaved
3. 03
   

   newLLM-Ready Training Dataset for Apple's Foundation Models (iOS 26)rileyhealth.gumroad.com

   ▲ 13rileygersh453d💬 5savesaved
4. 04
   

   newShow HN: Technical diagrams were slowing me down, so I built a faster waywww.rapidcharts.ai

   ▲ 5SamiCostox446ddiscusssavesaved
5. 05
   

   newI created a new training dataset for stale SOTA LLMs

   ▲ 2rileygersh449ddiscusssavesaved
6. 06
   

   newShow HN: Docs-to-Markdown – Convert Documentation URLs into Markdowngithub.com

   ▲ 2frandagostino598ddiscusssavesaved
7. 07
   

   newAsk HN: Has anyone seen this before in any ChatGPT?

   ▲ 1HereticGlaux474d💬 1savesaved
8. 08
   

   newShow HN: Open-sourced AI Agent runtime (YAML-first)github.com

   ▲ 1nsokra02216ddiscusssavesaved
9. 09
   

   newShow HN: Code+=AI – build AI webapps in minutes by having LLM complete ticketscodeplusequalsai.com

   ▲ 1cryptoz538ddiscusssavesaved
10. 10
    

    newShow HN: Interactive price comparison for AI models (LLMs and Speech-to-Text)ai-pricing.vercel.app

    ▲ 1nuxdie706ddiscusssavesaved
11. 11
    

    newShellGPT: A CLI productivity tool powered by LLMs like GPT-4github.com

    ▲ 3sea-gold678d💬 1savesaved
12. 12
    

    newShow HN: I open-sourced a $1M engine for closing loops in embedding spacegithub.com

    ▲ 8WFGY464d💬 10savesaved
13. 13
    

    newShow HN: PromptL, a templating language designed for LLM promptingpromptl.ai

    ▲ 7heycesr690d💬 4savesaved
14. 14
    

    newShow HN: Towards agentic Graph RAG: Enhancing graph retrieval with vector searchblog.kuzudb.com

    ▲ 6laminarflow027466ddiscusssavesaved
15. 15
    

    newShow HN: Lailaims – Talk to multiple LLMs at oncelailaims.pages.dev

    ▲ 6youz506ddiscusssavesaved
16. 16
    

    newShow HN: I built a tool to make system design diagrams less painfulwww.rapidcharts.ai

    ▲ 4SamiCostox447ddiscusssavesaved
17. 17
    

    newWould you use an LLM that follows instructions reliably?

    ▲ 3gdevaraj487d💬 5savesaved
18. 18
    

    newShow HN: LLM In-Browser Fuzzer Finds Hidden Prompt Injection in AI Browsersbrowsertotal.com

    ▲ 3minche353ddiscusssavesaved
19. 19
    

    newAccess LLM from Within Gsheet

    ▲ 2aidopper489d💬 2savesaved
20. 20
    

    newAsk HN: What if the AI scaling plateau is just a "false dip"?

    ▲ 1massicerro269d💬 3savesaved
21. 21
    

    newShow HN: Forge – 3MB Rust binary that coordinates multi-AI coding agents via MCPgithub.com

    ▲ 1vipdestiny238d💬 1savesaved
22. 22
    

    newShow HN: An Internationalization GitHub Action to Replace Crowdin with LLMsgithub.com

    ▲ 1cport1251d💬 1savesaved
23. 23
    

    newShow HN: Open LLM Spec – Standardizing inputs and outputs across providers

    ▲ 1gpt4o441d💬 1savesaved
24. 24
    

    newShow HN: Horizon – My AI-powered personal news aggregator and summarizergithub.com

    ▲ 1gg_equal_G225ddiscusssavesaved
25. 25
    

    newShow HN: Im making a tutorial Zero to LLM Agent; and it wrote its own agent loopwww.conseval.com

    ▲ 1macrolet225ddiscusssavesaved
26. 26
    

    newShow HN: Tako AI – Agent for Okta With Natural language (zero hallucination)github.com

    ▲ 1danFctr235ddiscusssavesaved
27. 27
    

    newShow HN: Dir2txt – Dump your project into clean LLM-ready text or JSONgithub.com

    ▲ 1shubhamoy535ddiscusssavesaved
28. 28
    

    newAsk HN: Would you use a pay-per-use AI chatbot with multiple models?

    ▲ 1husainnazer541ddiscusssavesaved
29. 29
    

    newDeepSeek v2.5 – open-source LLM comparable to GPT-4, but 95% less expensivewww.deepseek.com

    ▲ 193jchook704d💬 75savesaved
30. 30
    

    newShow HN: LLMRouter – Stop using GPT-4/o1 for everything (16 routing strategies)github.com

    ▲ 2tao2024279d💬 1savesaved

· edit