Generative AI On AWS Has Arrived With Amazon Bedrock (Console Overview and Demo)
HTML-код
- Опубликовано: 28 июн 2024
- Amazon Bedrock, AWS' new generative AI service, has just been released into general availability. Starting today, everyone can get access to the rich set of foundational models Bedrock provides and start integrating generative AI into their application using simple to use APIs. In this video, I walk you through the console experience of Bedrock and show you how to call a foundational model with code using Python.
Code: gist.github.com/beabetterdevv...
Bedrock Blog Post: aws.amazon.com/blogs/aws/amaz...
Sign up for my Newsletter to receive regular AWS updates AND get a FREE PDF that includes 5 AWS Project Ideas: beabetterdev.com/aws-project-...
📚 My Courses 📚
AWS Learning Accelerator - Learn AWS Through a Hands On Project - courses.beabetterdev.com/cour...
AWS Step Functions Masterclass - courses.beabetterdev.com/cour...
AWS Lambda - A Practical Guide - www.udemy.com/course/aws-lamb...
🎉SUPPORT BE A BETTER DEV🎉
Become a Patron: / beabetterdev
📚 INCREDIBLE BOOKS TO BUILD YOUR ABILITY TO FOCUS AND ORGANIZE 📚
Deep Work - amzn.to/463oPRk
So Good They Can't Ignore You - amzn.to/3PPo48N
Digital Minimalism - amzn.to/3Q00nep
A World Without Email - amzn.to/3t6nGKq
Time Block Planner - amzn.to/44Z9nEI
📚 MY RECOMMENDED READING LIST FOR SOFTWARE DEVELOPERS📚
Clean Code - amzn.to/37T7xdP
Clean Architecture - amzn.to/3sCEGCe
Head First Design Patterns - amzn.to/37WXAMy
Domain Driven Design - amzn.to/3aWSW2W
Code Complete - amzn.to/3ksQDrB
The Pragmatic Programmer - amzn.to/3uH4kaQ
Algorithms - amzn.to/3syvyP5
Working Effectively with Legacy Code - amzn.to/3kvMza7
Refactoring - amzn.to/3r6FQ8U
🎙 MY RECORDING EQUIPMENT 🎙
Shure SM58 Microphone - amzn.to/3r5Hrf9
Behringer UM2 Audio Interface - amzn.to/2MuEllM
XLR Cable - amzn.to/3uGyZFx
Acoustic Sound Absorbing Foam Panels - amzn.to/3ktIrY6
Desk Microphone Mount - amzn.to/3qXMVIO
Logitech C920s Webcam - amzn.to/303zGu9
Fujilm XS10 Camera - amzn.to/3uGa30E
Fujifilm XF 35mm F2 Lens - amzn.to/3rentPe
Neewer 2 Piece Studio Lights - amzn.to/3uyoa8p
💻 MY DESKTOP EQUIPMENT 💻
Dell 34 inch Ultrawide Monitor - amzn.to/2NJwph6
Autonomous ErgoChair 2 - bit.ly/2YzomEm
Autonomous SmartDesk 2 Standing Desk - bit.ly/2YzomEm
MX Master 3 Productivity Mouse - amzn.to/3aYwKVZ
Das Keyboard Prime 13 MX Brown Mechanical- amzn.to/3uH6VBF
Veikk A15 Drawing Tablet - amzn.to/3uBRWsN
🌎 Find me here:
Twitter - / beabetterdevv
Instagram - / beabetterdevv
Patreon - Donations help fund additional content - / beabetterdev
#aws
#bedrock
#llm
#ai
#generativeai
#amazonbedrock
This is great stuff. Thank you for the video.
Just a heads, Anthropic’s models have the largest context window (100k context or 75k words), and Anthropic is super concerned with who uses their models and that’s why the have the additional request requirements.
I’m still curious, why use them instead of a vector db that lets you add context or summarize then put in vector db - considering their model is not as good as OpenAI ?
Thank you so much!
Thanks.. great video
wow! thanks for the video!! BTW in Singapore, we can only use Anthropic model X( SAD
Thanks, for the Anthropic , fill in something you want (: works for me
Just so you know, it is "A (eye) 21" labs, not A121. Otherwise great video
I checked the bedrock aws pricing page but I did not saw clearly how much it would cost me for example to have custom models, I saw that for example Titan Text - Lite charges you 1.95 monthly for storing custom models in the fine-tuning section, but it charges you 6.40$ per hor if you use Provisioned Throughput. Could you explain me the different it got me a litte bit confused, thanks!
Just so if you are wondering why you are not able to see the providers and models, you need to make sure you are in the right region where these models are available. I had mine set to Ohio.. change it to N Virginia and now able to see the options..
Thank you. This was helpful.
Thanks for the tip!
Thank you
great stuff
Hi, First of all thanks for this video!
Question : will it be charging you for whatever it generated?
Is your avatar a digital human? Looks nice
What do I need to do if I want to deploy the API on a website, like as a chatbot? So basically what are the steps following adding the API to your code in order to deploy it? The user guide says for deploying agents, "configure your application to make API calls to that Agent alias." but that's pretty vague and I'm not sure how I would do that without a third party application.
nice intro. thx
Would be helpfull if you can explain us how much this costs and pricing
I just ran the script and got the error Unknown service: 'bedrock-runtime'. I was just thinking about this but if there are no aws credentials in this script how does it know to reach my aws account?
Can you please help me understand, bedrock lets you use more models but yet it costs me lots LOTS less to just use my OpenAI key and connect it to my lambda? Or am I missing something? Is the value here that you can use different models? It seems all models suck compared to latest openAI, or did I miss something ?
how does it know from the python code which account in AWS to reach out to? I dont see enviornment variables set up anywhere
In the lambda and endpoints
Do you not need to include an API key in the boto config or anything?
I would also like to know that
I just want to treat my collection of 1200 markdown files totaling about 2 million words, as one singular entity, as the entire collection is geared towards one project, I would like to interact with it and discuss semantics analysis, keeping track of themes and use artificial intelligence to maintain continuity over time, how on earth can I achieve this?
Did you make any progress with this? Most models seem to require you to add your own context on each request
@@mrbanana6969 no, I have been using Gemini 1.5 Early Access pro though, and the 1 million token limit does help, and I give it custom instructions first to account for what you said at the end there.
this is really cool but as he said approach with extreme caution. aws is great but the pricing stuff can truly, quickly sink you:/
techlead lolololol
The one and only.
Lol😂
The generated image doesn't look like Tech Lead at all... Are you saying that just because he's Asian?