OpenAI o1
Ranked 4 of 4 reasoning models by recommenders
People recommend o1 as an early step-change in AI reasoning on hard problems, though critics say its hype outran its real-world results and its price was high.
People recommend o1 as an early step-change in AI reasoning on hard problems, though critics say its hype outran its real-world results and its price was high.
22 people gave an opinion
- recommend it
- 12
- criticize it
- 10
- approval share
- 55%
Recommended in 7 of 13 months
- November 2025: recommended
- December 2025: recommended
- January 2026: recommended
- February 2026: not recommended
- March 2026: recommended
- April 2026: not recommended
- May 2026: recommended
- June 2026: not recommended
- July 2026: recommended
- August 2026: not recommended
- September 2026: recommended
- October 2026: not recommended
- November 2026: not recommended (month in progress)
Why it doesn’t carry the badge
- 12 people recommend itNeeds 10 or more, met.
- 1.2 recommenders for every criticNeeds 2 to 1 or better, not met.
- Recommended in 7 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Strong reasoning on hard questions
“also give the correct answer (without web search)”
Comment from 7 Jul 2026 in “A global workspace in language models” “OpenAI's o3, o1 and 4.1 get it right first try”
Comment from 7 Jul 2026 in “A global workspace in language models” “showed impressive reasoning capabilities”
Comment from 20 May 2026 in “An OpenAI model has disproved a central conjecture in discrete geometry” Major improvement over earlier models
“Last time there was such hype for a model was "strawberry" (o1) then gpt-5, and both turned out to be meaningful improvements but nowhere near the hype.”
Comment from 15 Nov 2025 in “A new Google model is nearly perfect on automated handwriting recognition” “around the time of o1/o3/Gemini 2.5 was when things changed and I haven't looked back since”
Comment from 25 Jan 2026 in “AI Tribalism”
What people warn about
Each mark is one comment
Hype exceeds real-world results
“Last time there was such hype for a model was "strawberry" (o1) then gpt-5, and both turned out to be meaningful improvements but nowhere near the hype.”
Comment from 15 Nov 2025 in “A new Google model is nearly perfect on automated handwriting recognition” “Hyped title. It was exclusively text-based diagnosis after physicians did the whole interview, exam, labs, etc.”
Comment from 4 May 2026 in “OpenAI's o1 correctly diagnosed 67% of ER patients vs. 50-55% by triage doctors” “it says the LLMs did not outperform any group of doctors in the most important cases”
Comment from 4 May 2026 in “OpenAI's o1 correctly diagnosed 67% of ER patients vs. 50-55% by triage doctors” Confident wrong answers persist
“This has been a visible problem at least since o1, and the lack of change tells me they do not have a real solution.”
Comment from 11 Dec 2025 in “GPT-5.2” “an LLM will happily leap to a conclusion and deliver it with high confidence”
Comment from 3 May 2026 in “OpenAI's o1 correctly diagnosed 67% of ER patients vs. 50-55% by triage doctors” “LLMs (o1) gets the following type of question wrong”
Comment from 16 Feb 2026 in “I want to wash my car. The car wash is 50 meters away. Should I walk or drive?” High cost for reasoning
“All thinking does is burn output tokens for accuracy, it is the AI getting high on its own supply”
Comment from 7 Apr 2026 in “System Card: Claude Mythos Preview [pdf]” “technically o1-pro is even more expensive, but I'm assuming that's a "please move on" pricing”
Comment from 20 Mar 2026 in “OpenCode – The open source AI coding agent” “Now, o1's price/performance seems like a distant bad dream.”
Comment from 14 Jan 2026 in “The Insecure Evangelism of LLM Maximalists”
Other reasoning models people recommend
The three most recommended other reasoning models, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| OpenAI o3 | 37 | 13 | 74% | 11 | — | Approved |
| VibeThinker 3B | 28 | 10 | 74% | 4 | — | Approved |
| DeepSeek R1 | 25 | 15 | 63% | 10 | — | — |
| OpenAI o1This page | 12 | 10 | 55% | 7 | — | — |
What people said
22 of the 23 comments that mention it, newest first. Each links to the original comment.
“o1 shows that OpenAI is both desperate and out of ideas.”
“also give the correct answer (without web search)”
“OpenAI's o3, o1 and 4.1 get it right first try”
“showed impressive reasoning capabilities”
“Hyped title. It was exclusively text-based diagnosis after physicians did the whole interview, exam, labs, etc.”
“it says the LLMs did not outperform any group of doctors in the most important cases”
“it's genuinely impressive AI can perform this level of success at a specific doctoring skill”
“If not it is just selling bait.”
“an LLM will happily leap to a conclusion and deliver it with high confidence”
“All thinking does is burn output tokens for accuracy, it is the AI getting high on its own supply”
“The early reasoning OpenAI models like O1 were really obvious about it (but also quite effective at convincing people the output was worthwhile”
“technically o1-pro is even more expensive, but I'm assuming that's a "please move on" pricing”
“this model was caught attempting to copy itself onto external servers”
“When openai trained a version of o1 on ARC it got worse at everything else.”
“LLMs (o1) gets the following type of question wrong”
“around the time of o1/o3/Gemini 2.5 was when things changed and I haven't looked back since”
“Now, o1's price/performance seems like a distant bad dream.”
“Are these models at that level yet? Or are they o1?”
“This has been a visible problem at least since o1, and the lack of change tells me they do not have a real solution.”
“The amount of compute you could throw at a problem was absurd.”
“SOTA 'Reasoning Models' (o1/GPT-4) are actually unusable for agent swarms”
“Last time there was such hype for a model was "strawberry" (o1) then gpt-5, and both turned out to be meaningful improvements but nowhere near the hype.”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.