Hackernews posts about GPT3

GPT3 is a revolutionary AI model that uses machine learning to generate human-like text based on input prompts, allowing for applications such as language translation, content creation, and natural language processing.

1. Three Years from GPT-3 to Gemini 3 (www.oneusefulthing.org)
   352 points by JumpCrisscross 12 days ago | 319 comments
2. Three Years from GPT-3 to Gemini 3 (www.oneusefulthing.org)
   4 points by dkobia 16 days ago | discuss
3. LLMs let me maintain my PostgreSQL extension for PRQL after becoming a parent
   5 points by kaspermarstal 25 days ago | discuss
4. Show HN: I built CostLens SDK to cut my AI bills by routing to cheaper models (costlens.dev)
   2 points by j_filipe 18 days ago | 2 comments
5. Show HN: Investigating why GPT-5 has made ChatGPT 'broken' (muhammadasmulkana.substack.com)
   1 points by muhammad-shafat 14 days ago | discuss
6. Don't push AI down our throats (gpt3experiments.substack.com)
   430 points by nutanc 4 days ago | 356 comments
7. OpenAI is facing every startup's VC question: What if Google copies you? (gpt3experiments.substack.com)
   1 points by nutanc 1 day ago | discuss
8. GPT-5.1: A smarter, more conversational ChatGPT (openai.com)
   555 points by tedsanders 22 days ago | 726 comments
9. Building more with GPT-5.1-Codex-Max (openai.com)
   483 points by hansonw 15 days ago | 319 comments
10. Reverse engineering Codex CLI to get GPT-5-Codex-Mini to draw me a pelican (simonwillison.net)
    168 points by simonw 26 days ago | 76 comments
11. GPT-5.1 for Developers (openai.com)
    112 points by tedsanders 21 days ago | 29 comments
12. GPT-5-Codex-Mini – A more compact and cost-efficient version of GPT-5-Codex (github.com)
    56 points by wahnfrieden 26 days ago | 54 comments
13. GPT-OSS 120B Runs at 3000 tokens/sec on Cerebras (www.cerebras.ai)
    48 points by samspenc 27 days ago | 29 comments
14. Show HN: A business SIM where humans beat GPT-5 by 9.8 X
    23 points by sumit_psp 15 days ago | 14 comments
15. Early experiments in accelerating science with GPT-5 (openai.com)
    18 points by tabletcorry 14 days ago | discuss
16. How we helped a YC company (Upsolve) catch a GPT-5 regression (www.arthur.ai)
    17 points by fryz 20 days ago | discuss
17. VLM Showdown: GPT vs. Gemini vs. Claude vs. Orion (chat.vlm.run)
    15 points by fzysingularity 14 days ago | 1 comments
18. Can GPT-5 Beat My Favorite Daily Puzzle Game? (www.nicksypteras.com)
    10 points by nsypteras 20 days ago | 4 comments
19. Baidu releases open-source multimodal AI that it claims beats GPT-5 and Gemini (venturebeat.com)
    9 points by teleforce 22 days ago | 6 comments
20. GPT-5-Thinking using Grokipedia as a source (www.reddit.com)
    5 points by timpera about 13 hours ago | 2 comments
21. Early science acceleration experiments with GPT-5 [pdf] (cdn.openai.com)
    5 points by gronky_ 14 days ago | 1 comments
22. OpenAI readies GPT-5.1 Thinking model ahead of Gemini 3 Pro (www.testingcatalog.com)
    5 points by gmays 26 days ago | 1 comments
23. Wall Street Experts Tested GPT-5 and Claude. Both Struggled – Even with Excel (surgehq.ai)
    5 points by holdingunsteady 28 days ago | 1 comments
24. Using Codex CLI with GPT-OSS:120B on an Nvidia DGX Spark via Tailscale (til.simonwillison.net)
    5 points by GavinAnderegg 28 days ago | discuss
25. Early experiments in accelerating science with GPT-5 (openai.com)
    4 points by sanjitb 11 days ago | 1 comments
26. Early science acceleration experiments with GPT-5 (arxiv.org)
    4 points by Anon84 12 days ago | 1 comments
27. Some context on why some 80s kids keep getting mistaken for GPT (old.reddit.com)
    4 points by neilv 19 days ago | 1 comments
28. OpenAI releases GPT-5.1 alongside eight new ChatGPT personality styles (arstechnica.com)
    4 points by moelf 22 days ago | 1 comments
29. GPT-image-1 prompted to "create the same image" 74 times (old.reddit.com)
    4 points by SweetSoftPillow 23 days ago | 1 comments
30. Gemini 3 vs. GPT 5.1 for RAG (agentset.ai)
    4 points by tifa2up 15 days ago | discuss