Hackernews posts about GPT-3.5

1. Show HN: Open-source study to measure end user satisfaction levels with LLMs (open-llm-initiative.com)
   12 points by sparacha 23 days ago | 2 comments
2. Show HN: Cursor Search – AI-powered search that follows your cursor (cursorsearch.com)
   9 points by kadarla 24 days ago | 3 comments
3. Fine-tune your own Llama 2 to replace GPT-3.5/4
   955 points by kcorbitt about 1 year ago | 182 comments
4. Phind Model beats GPT-4 at coding, with GPT-3.5 speed and 16k context (www.phind.com)
   891 points by rushingcreek 11 months ago | 347 comments
5. Gpt4all: A chatbot trained on ~800k GPT-3.5-Turbo Generations based on LLaMa (github.com)
   593 points by qeternity over 1 year ago | 301 comments
6. GPT-3.5 crashes when it thinks about useRalativeImagePath too much (iter.ca)
   465 points by goranmoomin 8 months ago | 151 comments
7. GPT-3.5 Turbo fine-tuning and API updates (openai.com)
   395 points by davidbarker about 1 year ago | 233 comments
8. Smarter summaries with finetuning GPT-3.5 and chain of density (jxnl.github.io)
   234 points by ivanleomk 10 months ago | 39 comments
9. Microsoft Teams Premium: powered by OpenAI’s GPT-3.5 (www.microsoft.com)
   215 points by cloudking over 1 year ago | 159 comments
10. Fine-tuning GPT-3.5-turbo for natural language to SQL (medium.com)
    208 points by saigal about 1 year ago | 75 comments
11. GPT4 is up to 6 times more expensive than GPT3.5 (openai.com)
    192 points by behnamoh over 1 year ago | 131 comments
12. GPT-4, without specialized training, beat a GPT-3.5 class model that cost $10M (www.threads.net)
    161 points by CharlesW 6 months ago | 127 comments
13. Pi.ai LLM Outperforms Palm/GPT3.5 (inflection.ai)
    151 points by ergodas about 1 year ago | 86 comments
14. 1960s chatbot ELIZA beat OpenAI's GPT-3.5 in a recent Turing test study (arstechnica.com)
    136 points by layer8 10 months ago | 47 comments
15. Show HN: Jailbreaking GPT3.5 Using GPT4 (github.com)
    134 points by raghavtoshniwal over 1 year ago | 39 comments
16. Can GPT-4 and GPT-3.5 play Wordle? (twitter.com)
    117 points by amichail over 1 year ago | 104 comments
17. GPT-3.5 and Wolfram Alpha via LangChain (huggingface.co)
    107 points by heliophobicdude over 1 year ago | 8 comments
18. Dust XP1 switches to GPT-3.5-turbo, is now free to use (dust.tt)
    99 points by ukuina over 1 year ago | 50 comments
19. OpenAI Releases Function Calling for GPT-3.5 & GPT-4 (platform.openai.com)
    87 points by nlh over 1 year ago | 2 comments
20. On the left is GPT-3.5. On the right is GPT-4 (twitter.com)
    78 points by susam over 1 year ago | 93 comments
21. Benchmarking the accuracy of GPT3.5's and GPT-4's code generation abilities (github.com)
    74 points by seanmor5 over 1 year ago | 29 comments
22. We gave GPT-3.5 tools to run, write, commit, and deploy code (old.reddit.com)
    71 points by satvikpendem over 1 year ago | 33 comments
23. Microsoft says GPT 3.5 has 20B parameters? (arxiv.org)
    57 points by Heidaradar 11 months ago | 25 comments
24. Ask HN: Is GPT 4's quality lately worst than GPT 3.5?
    51 points by agonz253 about 1 year ago | 40 comments
25. GPT 3.5 vs. Llama 2 fine-tuning: A Comprehensive Comparison (ragntune.com)
    47 points by samlhuillier about 1 year ago | 12 comments
26. GPT-3.5 passed yet another Theory of Mind test (twitter.com)
    40 points by izzygonzalez over 1 year ago | 50 comments
27. ChatGPT recreated with GPT-3.5? (github.com)
    40 points by karfly over 1 year ago | 18 comments
28. Meta unveils Llama 2 Long AI beats GPT-3.5 Turbo and Claude 2 on some tasks (venturebeat.com)
    39 points by nickthegreek 12 months ago | 12 comments
29. A GPT 3.5 powered tool for generating regex (fuckregex.dev)
    38 points by DefineOutside over 1 year ago | 23 comments
30. GPT-3.5-turbo-instruct can play chess around 1800 Elo (twitter.com)
    38 points by beefman about 1 year ago | 15 comments
31. GPT4All outscores GPT-3.5 on new hallucination metric (github.com)
    36 points by manyoso over 1 year ago | 13 comments
32. Gpt3.5 just adding a random dude's photo in the reply (old.reddit.com)
    32 points by decide1000 11 months ago | 8 comments