Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
achierius
8 months ago
|
parent
|
context
|
favorite
| on:
OpenAI o3 and o4-mini
How is this a notable release? It's strictly worse than Gemini 2.5 on coding &c, and only an iterative improvement over their own models. The only thing that struck me as particularly interesting was the native visual reasoning.
famouswaffles
8 months ago
[–]
It's not worse on coding. SWE Bench, Aider, live bench coding all show noticeably better results.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: