LoRAs are weaving their way into SD3.5 already
OPEN SOURCE
-
Open-source AI vs concentrated power and dangerous applications regulation
By
–
IMO, this is a misleading analogy. "AI" is more akin to "physics" than to "nuclear bombs". You want AI open-source and transparent for maximum progress and minimum concentration of power and then regulate/limit the dangerous applications of it (like AI for nuclear bomb, AI for
-
Open Source AI Delivers Significant Societal Benefits
By
–
open source ai creates incredible societal benefit
-
Open Table Formats and Interoperability with Delta Lake and Iceberg
By
–
Join us for a discussion on interoperability with Michael Armbrust, creator of Delta Lake, and Ryan Blue, co-creator of Apache Iceberg! They explore open table formats — and how Databricks is tackling interoperability challenges: https://
dbricks.co/4ejg3my -
KlingAI Virtual Try-On GitHub Repository
By
–
github: https://
github.com/AtaUllahB/Klin
gAI-Virtual-Try-On
… -
LangChain Celebrates Two Years of Open Source Innovation
By
–
LangChain turns two! Forever grateful to everyone who has been a part of this journey LangChain has changed a TON from the open source side project it started as. I wanted to take this time to reflect on the evolution Thank you for being a part
-
Flowise v2.1.3 Releases Advanced RAG File Upload Features
By
–
Flowise v2.1.3 is now out and comes with a lot of goodies🎃
— FlowiseAI (@FlowiseAI) 24 octobre 2024
📂Full and RAG File Uploads
– RAG file uploads embed and upsert to the vector store in real-time.
– Full file uploads send entire files to the LLM, ideal for large-context models like Gemini and Claude.
See how we… pic.twitter.com/zLuzJWJv9RFlowise v2.1.3 is now out and comes with a lot of goodies Full and RAG File Uploads
– RAG file uploads embed and upsert to the vector store in real-time.
– Full file uploads send entire files to the LLM, ideal for large-context models like Gemini and Claude. See how we -
Llama 3.2 Quantized Versions Speed Up Inference 2-4x
By
–
We want to make it easier for more people to build with Llama — so today we’re releasing new quantized versions of Llama 3.2 1B & 3B that deliver up to 2-4x increases in inference speed and, on average, 56% reduction in model size, and 41% reduction in memory footprint.
Details -
ORPO Guide: The Only Resource Covering Pruning Techniques
By
–
And, of course, don't forget the best guide about ORPO. The other resources don't even mention pruning.
-
MobileNetV3 Image Classification Community Template
By
–
Community Template 3: Image Classification with MobileNetV3 from @jsonmathsai (
https://
github.com/SakanaAI/AI-Sc
ientist/pull/141
…)