Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
Mkengin
20 days ago
|
parent
|
context
|
favorite
| on:
GPT-5.2-Codex
According to SWE-Rebench Anthropic and OpenAI are really close in performance, while GPT-5.2 costs less than half the cost of CC per problem.
https://swe-rebench.com/
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
https://swe-rebench.com/