GPT
OpenAI's family of large language models, available through ChatGPT and the OpenAI API.
HN commenters like GPT most as a code reviewer and plan checker alongside other models,1234 and for following instructions closely without refusing.56 The most common complaint is verbose or overengineered output,789 and some say it rarely explains its reasoning or assumptions.1011
Why people recommend it
Code review · 3
Often used to review plans and check other models' code, catching gaps and edge cases.
“I use GPT only for audits and serious code logic because it can catch gaps and inconsistencies that Claude misses.”
“GPT works well as a validator. It tends to consider far more edge cases than I would.”
Show 1 more
“then gpt review the plan and provide feedback”
Core logic · 3
Seen as more thorough on core logic and architecture than some alternatives.
“GPT series models are more thorough and better”
“The GPTs are better at general architecting”
Show 1 more
“It appears both Claude and GPT believe GPT's solutions are better (and so I do).”
Instruction following · 3
Praised for sticking to user rules and doing work other models refuse.
“This (among many others) is the reason I use GPT over Claude. It adheres very closely to your instructions and rules.”
“Yes, I tend to prefer GPT”
Show 1 more
“I recently switched to using GPT for Nelson and the security benchmarks”
Tone · 3
Some prefer its plainer, less confrontational tone over Claude's.
“OpenAI models are much more tame and to the point (I use the same prompt)”
“I just switched to the GPT models and it's a breath of fresh air”
Show 1 more
Plan limits · 2
The $20/mo plan is described as enough for daily work.
“GPT is inexhaustible on the $20/mo plan”
“I use GPT for months. $20/m plus plan is enough for my daily work.”
Explaining concepts · 2
Used to simplify complex material and find learning resources.
“Take it to sonnet 5 or gpt and ask it to explain this to a layman.”
“Just ask GPT or Claude to find some good courses and learning paths on GitHub.”
Caveats and dissent
Verbose output · 3
Code and text can come out overengineered, overly defensive or padded.
“GPT's code ends up way more verbose/complex/overengineered than it needs to be”
“GPT works well as a validator. It tends to consider far more edge cases than I would.”
Show 1 more
“Claude/GPT really wants to append more text for some reason.”
Hidden reasoning · 2
Some say it delivers without sharing assumptions or thinking.
“I value gpt so much, but it is such a worse peer to me than the other models I use.”
“It was so jarring to switch to GPT and Opus and find that they won't discuss with me, won't reveal their thinking”
Accuracy errors · 2
Users report repeated arithmetic mistakes and poor results in Elixir.
“GPT messed up the math for me the other day when I was simply adding 10 durations for a TRT.”
“I have found GPT to be pretty awful at writing Elixir”
Prose writing · 1
One commenter finds it unacceptable for writing English text.
“Neither Claude nor GPT are acceptable for writing English text.”
Refusals · 1
One user reports it refused a task it considered illegal.
“GPT refused to do so (citing that it's illegal even though I own the games).”
Often compared with
3 moved from Claude to GPT
1 moved from GPT to Claude
1 moved from Claude Opus to GPT
1 moved from GPT to Claude Opus
1 moved from GPT to GeminiGLM9 comments mention both
Who recommends it in their own words
Threads where it came up
- Path to Astra: critical capabilities and frontier safeguards 2 mentions
- Analyzing student votes across AI models for college essay help 2 mentions
- Ask HN: How much money do you spend monthly on subscriptions for AI models? 2 mentions
- Fable 5 vs. GPT-5.6 Sol on an NP-Hard Problem: Does /goal help? 2 mentions
- When I reject AI code even if it works 2 mentions
- Where AI agents' social boundaries break: 5 LLMs try to form a government 1 mention
Data: gpt.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 47
“GPT for the images of the characters”
“GPT for the images of the characters”
“I did ask GPT to port over my key commands and shortcuts, which it did, and things are great again!”
“then I’d throw it into GPT and post the answer”
“GPT messed up the math for me the other day when I was simply adding 10 durations for a TRT.”
“I use GPT, Codex, Claude, Cursor and other current tools extensively.”
“I prefer talking to GPT”
“I pay for GPT, Claude and Gemini.”
“I just switched to the GPT models and it's a breath of fresh air”
“I find GPT to be better at understanding physics conceptually but Claude is better at writing numerical code.”
“Claude/GPT really wants to append more text for some reason.”
“that's how exactly I use GPT and Claude in China”
“Sticking to a mix of GPT and Gemini for now.”
“I find it hard to distinguish between the outputs of GPT/Claude/Kimi/GLM recently”
“I'm using GPT (Terra, Sol, Luna), Claude (Opus 5, Fable), Qwen 3.8 and GLM 5.3 Flash daily”
“I eventually asked gpt if I was correct in my understanding about refrigeration controllers”
“feeding them through GPT and Claude one by one, and it's drastically increased the speed at which I can knock out a completed piece”
“Four of the eight clubs are managed by frontier AI models (Claude, GPT, Gemini, Manus).”
“I use GPT for planning/roadmaps and code review, it checks Fable.”
“GPT gets the most out there, but creates interesting nuance that turn an idea into a more creative exercise.”
“I asked GPT and it suggested that I use ficl”
“I was actually using it more with GPT models but they have been getting better.”
“Just ask GPT or Claude to find some good courses and learning paths on GitHub.”
“It appears both Claude and GPT believe GPT's solutions are better (and so I do).”
“I've seen the same with GPT models, with Grok, with Deepseek”
“my usual GPT/Claude/DeepSeek for code”
“I do also use the API's for GPT and Gemini for some side projects”
“I naturally use several models—GPT, Claude, Gemini, Grok, Perplexity, etc.”
“I use Codex, gpt, and Claude in various configurations”
“I value gpt so much, but it is such a worse peer to me than the other models I use.”
“The GPTs are better at general architecting”
“This (among many others) is the reason I use GPT over Claude. It adheres very closely to your instructions and rules.”
“Neither Claude nor GPT are acceptable for writing English text.”
“you can just pay as you go with GPT and Fable”
“I use GPT only for audits and serious code logic because it can catch gaps and inconsistencies that Claude misses.”
“I have found GPT to be pretty awful at writing Elixir”
“Even just spending 10 minutes forcing GPT to write a design doc results in much more robust plans than plan mode”
“I do about 10 google search queries for every 1 opus/gpt prompt”
“using Claude/GPT to ask for references etc is good enough for me”
“Take it to sonnet 5 or gpt and ask it to explain this to a layman.”
“Yes, I tend to prefer GPT”
“GPT's code ends up way more verbose/complex/overengineered than it needs to be”
“It was so jarring to switch to GPT and Opus and find that they won't discuss with me, won't reveal their thinking”
“then gpt review the plan and provide feedback”
“I use 3 AI's (Claude, GPT and Gemini) to review each other's design plans and implementation on the same code base.”
“using GPT purely as an adversarial collaborator to spot check Claude's work”
“GPT works well as a validator. It tends to consider far more edge cases than I would.”