Good question We are running chat-langchain in production: https://
github.com/langchain-ai/c
hat-langchain
… Includes an incremental ingest pipeline run daily, citing of source quotes, returning retrieved documents, query transformation, chat memory, and streaming
TOOLS
-
Chat-LangChain Production: RAG Pipeline with Daily Ingestion
By
–
-
Pika Art Transforms Images into Videos with AI
By
–
Image-to-video: It’s like giving any picture a movie makeover. 📽️
— Pika (@pika_labs) 22 décembre 2023
Sign up at https://t.co/nqzjGy82Lx pic.twitter.com/P6WCjDqjC1Image-to-video: It’s like giving any picture a movie makeover. Sign up at http://
pika.art -
Ambient Motion Feature Launched for Motion Brush Gen-2
By
–
Paint an area. Set the intensity. Watch your image come to life.
— Runway (@runwayml) 22 décembre 2023
Ambient Motion now available for Motion Brush in Gen-2. pic.twitter.com/AQS4kaViSkPaint an area. Set the intensity. Watch your image come to life. Ambient Motion now available for Motion Brush in Gen-2.
-
Choosing the Best LLM Framework: LlamaIndex, LangChain, OpenAI Assistants, or Custom
By
–
In this one, we help you demystify the role of the most popular options for building your LLM application: @llama_index
, @langchain
, @OpenAI Assistants, and building everything from scratch. But most importantly, we guide you in making the best choice for your own application. -
RAG and LLMs Course Finale: LangChain vs LlamaIndex vs OpenAI
By
–
Good morning fellow AI enthusiast! This week's iteration focuses on the last video of our 3-course series on RAG and LLMs with @towards_AI and @activeloop
! https://
open.substack.com/pub/louisbouch
ard/p/langchain-vs-llamaindex-vs-openai?r=25qlky&utm_campaign=post&utm_medium=web&showWelcome=true
… -
AI Image Upscaler Script Using ControlNet Tile Model
By
–
I've made a little script for upscaling any image using Replicate and a controlnet-tile model – for high quality upscales.
— fofr (@fofrAI) 22 décembre 2023
Just run: upscale "example.jpg"
Upscaling takes about 30s
You can change all the params from the command line.
Links and examples👇 pic.twitter.com/mZLFB5hWB7I've made a little script for upscaling any image using Replicate and a controlnet-tile model – for high quality upscales. Just run: upscale "example.jpg"
Upscaling takes about 30s You can change all the params from the command line. Links and examples -
Baichuan Upgrades AI Model with Enhanced Document Processing
By
–
Furthermore, Baichuan has also upgraded the model experience on our official website, now supporting various text uploads, including PDFs, Word documents, and URL inputs.
Click here to get started: https://
baichuan-ai.com -
Baichuan AI Launches Turbo-192K API with Search Augmentation
By
–
On December 19th, Baichuan AI launched the Baichuan2-Turbo series API based on Search Augmented capabilities, which includes Baichuan2-Turbo-192K and Baichuan2-Turbo. It supports 192K context window and enhances the search capabilities of knowledge base. https://
platform.baichuan-ai.com/playground -
MidJourney V6 Review: Contrast, Style and Hand Generation Issues
By
–
I've been under the weather today and haven't accomplished much but I did get a chance to spend some time with MidJourney V6. Here's some things that I noted… I'm loving the contrast and style of the images but it still has issues from time to time with hands… It's gotten
-
Midjourney v6 upscaling review compared to Magnific
By
–
Here's an original image that I created: And here's the same image Upscaled (Creative): The difference isn't massive. It's not like Magnific where you'll see a lot of whole new details added in (yet). I'm super excited about MJ6 but it doesn't feel to me as big of a leap that