Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
zargon
16 days ago
|
parent
|
context
|
favorite
| on:
Accelerating Gemma 4: faster inference with multi-...
Groq.
beavisringdin
16 days ago
[–]
No, it was a custom ASIC chip with weights baked in for a singular model. I do envision a future where we return to cartridges. Local AI is de facto and massively optimised chips are built to be plug and play running a single SoTA model.
SJMG
16 days ago
|
parent
[–]
Likely
https://taalas.com
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: