Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
free_bip
4 months ago
|
parent
|
context
|
favorite
| on:
Microsoft Phi-3 Cookbook
Looking at the benchmarks, it seems like Phi-3 Small (7B) marginally beats out Llama3-8B on most tasks, which is pretty exciting!
zone411
4 months ago
|
next
[–]
On my benchmark (NYT Connections), Phi-3 Small performs well (8.4) but Llama 3 8B Instruct is still better (12.3). Phi-3 Medium 4k is disappointing and often fails to properly follow the output format.
Filligree
4 months ago
|
prev
|
next
[–]
Have you found either model to be good enough to do anything interesting, reliably?
ukuina
4 months ago
|
parent
|
next
[–]
Llama3-8B is adequate for non-technical summarization or simple categorization.
ashu1461
4 months ago
|
prev
[–]
It also seems to be comparable to gpt 3.5 turbo which I feel hard to believe. People have obviously found out a way around these benchmarks.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: