Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
jerrygenser
on March 12, 2025
|
parent
|
context
|
favorite
| on:
Gemma 3 Technical Report [pdf]
I suppose siglipv2 wasn't out yet when they were training this - I wonder if there will be an update to the multimodal models or pali-gemma which utilizes Siglip2. Aya Vision from Cohere utilized siglip 2 to great effect
changtimwu
on March 13, 2025
[–]
For llama.cpp, this would involve another handcrafted Siglip2 NN process. Interesting—they just figured out how to handle Siglip1.
https://github.com/ggml-org/llama.cpp/pull/12344/commits/631...
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
