Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
kseniamorph
5 months ago
|
parent
|
context
|
favorite
| on:
GPT‑5.4 Mini and Nano
wow, not bad result on the computer use benchmark for the mini model. for example, Claude Sonnet 4.6 shows 72.5%, almost on par with GPT-5.4 mini (72.1%). but sonnet costs 4x more on input and 3x more on output
PunchTornado
5 months ago
[–]
what's the point of this benchmark if sonnet is working great at my tasks and mini can't solve my tasks?
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
