GPT-3
Fewer than 5 people recommend it so far, so there's no written verdict. Here is what commenters said.
Why people recommend it
“I've been getting good programming and molecular biology results out of these back to GPT3.5.”
“GPT-3, which was a pretty cool text generator model”
Show 1 more
“Frankly, I saw more creativity from plain GPT 3.”
Caveats and dissent
“they did advertise gpt-3.5-turbo as a straight up improvement”
“all but 3.5 Turbo could be convinced”
Show 1 more
“Is offering GPT 3.5 Turbo and Gemini 1.5 Pro.”
By version sentiment of people who named that release, oldest release first
Versions with at least 5 people. 0 people didn't name a version. Small samples swing a lot, so read the counts too. Compare all models →
Often compared with
Threads where it came up
- Sycophancy is the first LLM "dark pattern" 2 mentions
- Vote on which of Hacker News' challenges for AI have been met 1 mention
- Pnpm 12.0 1 mention
- Accelerating GPT-5.6 Sol Ultrafast 1 mention
- Agentic Coding Is a Trap 1 mention
- Codex for Almost Everything 1 mention
Also written as GPT-3, GPT-3.5 Turbo, gpt-3.5-turbo. Data: gpt-3.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 16
“GPT 3.5 or something used to be surprisingly good at chess”
“With 3.5 Turbo I used to give it bulleted information and paragraphs of what I wanted it to sound like and it'd replicate that style”
“they did advertise gpt-3.5-turbo as a straight up improvement”
Great article! I use LLMs 99.9% in the chat box, to bounce ideas around, ask research and reference questions, and so forth. However, I have sworn off 'agents' entirely for serious code, for reasons your article captures almost totally and perfectly. I'll still use 'agents' for throwaway tasks--mostly with local models--including tasks where some sort of ad-hoc code generation is in the critical path (e.g. scraping d
“I use older models mostly chatgpt3.5”
“I've been getting good programming and molecular biology results out of these back to GPT3.5.”
“all but 3.5 Turbo could be convinced”
“GPT-3, which was a pretty cool text generator model”
/shrug I have no intention of changing your mind. I don't think of the people I reply to highly enough to believe they can change their minds. I reply to these comments for other people to read. Think of it as me adding ky point of view for neutral readers. Either way, I could use AI for some coding tasks back in GPT 3.5 days. It was unreliable, but not completely useless (far from it in fact) Nowadays it is a little
“Frankly, I saw more creativity from plain GPT 3.”
“going all the way back to GPT3 to do codegen in the Playground”
“GPT3 was originally just a completion model.”
“the same way we used GPT-3”
“Is offering GPT 3.5 Turbo and Gemini 1.5 Pro.”
“albeit not as good”
“I think it even worked with GPT-3.5 for trivial things.”