GENERATIVE AI - AI Dynamics

The book is GPT-2 based, which is more beginner friendly. The Llama 3 code is bonus material. (It makes sense to implement GPT-2 first and than convert that to Llama 3 imho.) Llama 3 is dense, not an MoE though.

→ View original post on X — @rasbt

1 January 2025

Inference Performance Not Bad For Extended Usage

By

@localghost

–

01 January 2025 22h35

it’s not bad, you won’t be constantly inferencing for hours

→ View original post on X — @localghost

1 January 2025

Top Tech Trends 2025: AI Spending, Agents, and Space Economy

By

@kirkdborne

–

01 January 2025 21h27

Get the 2025 @CBinsights Top Tech Trends Report here: https://
cbinsights.com/research/repor
t/top-tech-trends-2025/
… Attached charts from @CBinsights = 4 hot trends from 2024:
1. AI’s cash grab
2. Big tech’s #AI spending spree
3. AI Agent landscape explodes
4. The new space economy

→ View original post on X — @kirkdborne

1 January 2025

AI Image Models Should Be Tools Not Artists Says User

By

@soumithchintala

–

01 January 2025 20h55

what I'm finding is that, the models want to be more of an artist than a replacement for photoshop — which is fine, but I want to be the artist here, and want the tool to be more of a "magically easier photoshop where I ask it what to do in detail, and it does that — not more

→ View original post on X — @soumithchintala

1 January 2025

Image Generation Models Struggle With Precise Localized Edits

By

@soumithchintala

–

01 January 2025 20h54

i'll give a representative but not exact example: change the color of X's shirt from blue to red: the generations often change the entire shirt style itself — they don't respect how much and what I'm trying to change, and dont try to preserve details I ask to preserve

→ View original post on X — @soumithchintala

1 January 2025

AI Image Transformation Tools: Selective Detail Preservation

By

@soumithchintala

–

01 January 2025 20h46

what are AI products that allow me to transform existing images, while preserving some selective details (that i select), like faces, areas, etc.? the tools I've used so far only take the selection as a hint, or dont generate well around the selection? trying personalized art

→ View original post on X — @soumithchintala

1 January 2025