21 сағат бұрын
21 күн бұрын
21 күн бұрын
21 күн бұрын
Ай бұрын
Ай бұрын
4 ай бұрын
Пікірлер
@BradPitson 16 сағат бұрын
Hi, do you have python code to work with azure ai search
@DataStaxDevs 5 сағат бұрын
That's a great question for our Discord channel! discord.com/invite/datastax Head on over and the team will be happy to chat.
@ErickWendelAcademy 7 күн бұрын
🥳🥳🥳🥳🥳
@123jamalq 8 күн бұрын
Excellent content, Can i do the same with KZbin video....please make a vodeo...thanks.
@DataStaxDevs 7 күн бұрын
Sounds like a good question for our dev rel team over on Discord: discord.com/invite/datastax
@123jamalq 8 күн бұрын
Hi, This is great. Can i do the same with any youtube video. Please make a video on this.
@kallolbecs 8 күн бұрын
how is it any different from other existing solutions like vectorshift?
@DataStaxDevs 7 күн бұрын
Langflow simplifies building AI workflows with a low-code interface, while Vectorshift specializes in managing and searching vector embeddings for AI-driven insights. If you want to know more, our dev rel team is ready to meet you on Discord! discord.com/invite/datastax
@roccov1972 9 күн бұрын
Thanks for the great video! I’m new to the whole 'Lang' world, and your explanation of LangFlow was super helpful and easy to follow. It’s exciting to see how it simplifies workflows-I can’t wait to explore it more!
@DataStaxDevs 8 күн бұрын
Great to hear!
@JeremyDevz 9 күн бұрын
Awesome!! Removing the friction. Thank you DataStax 🙌🏽🙌🏽
@DataStaxDevs 8 күн бұрын
🤩
@HazemAzim 9 күн бұрын
SearchGPT is out there
@popeefrog4358 9 күн бұрын
How is .new different from what you have already and did you release a pricing plan for Langflow?
@DataStaxDevs 7 күн бұрын
Langflow.new lets you try out Langflow without signing up for the service. Langflow is free to use as an open-source tool. If you're using it with cloud services or other DataStax products (e.g., Astra DB), there may be associated costs for those services. You can learn more about DataStax pricing here: www.datastax.com/pricing/astra-db
@raedkm 11 күн бұрын
at least give us 4k videos.
@DataStaxDevs 11 күн бұрын
thanks for the feedback!
@yanweifuture 13 күн бұрын
When I click the "Model Name Refresh button", an error occurs. The error message is "Could not retrieve models. Please, make sure Ollama is running." I installed Ollama on my Windows PC, and Langflow runs in Docker. So please kindly help me if you have any ideas about that.
@yanweifuture 13 күн бұрын
The Ollama server works normally; I can use it in a command-line window.
@yanweifuture 13 күн бұрын
I got that. Set the base URL as "host.docker.internal:11434". But I still don`t know how to set a directory path.
@davidgilardi6923 13 күн бұрын
I have a hunch it's because you are running Langflow from within a Docker container and it's not getting the correct access back out to your Windows OS or possibly the URL/port combo is incorrect. I haven't tried this exact setup myself as I am running Langflow from my Linux subsystem in Mac via a virtual environment. You might try gaining access to Ollama via your Docker container shell first.
@thehard-coder9398 14 күн бұрын
Hi sir - I got my ollma and model up and running my local machine. However, when I drag an Ollama component to the LF's canvas I saw a message "No Parameters are available for display" on the model list. Any idea on this? I would appreciate your response. Thanks
@davidgilardi6923 14 күн бұрын
Hi @thehard-coder9398. Where is your Langflow instance running? Are you running it locally or using the hosted version on DataStax? My initial hunch is that you are running Langflow somewhere that does not have immediate access to your local Ollama instance running from your localhost. Notice the default setting for the BaseURL of localhost:11434. You can do this from a hosted version, but it takes a little more setup because you need to host Ollama somewhere that internet based cloud instances can access. Matter of fact, we are currently working on a video on this very topic. For now, my advice, if you're not already, would be to install Langflow locally and try there unless you're already familiar with deploying a service like Ollama to the cloud.
@thehard-coder9398 13 күн бұрын
@@davidgilardi6923 - Thank you for your prompt reply, sir. Indeed it was on Cloud based. I shall do it locally then. Million thanks!
@marcosoliveira8731 15 күн бұрын
Really useful information very well delivered!
@DataStaxDevs 14 күн бұрын
Thanks for tuning in!
@TheHallofFameVault 18 күн бұрын
Great! Can we have multiple outputs from the flow using the structured output feature i.e, while a user continues to chat with the chatoutput, the structured output feature collects specific data and stores it in a db or specified files
@davidgilardi6923 18 күн бұрын
Great question. I don't see why not. You could definitely wire up the "chat" portion to the ChatOutput for conversation with your user while having a parallel connection to a structured output component. The "Response" node from your agent supports multiple connections. I just tested this locally using multiple structured outputs, each with their own params, and connected to a vector DB collection using the Astra DB component along with the regular chat output. I get the conversation and all of the user metadata is stored in my vector DB as metadata.
@TheHallofFameVault 15 күн бұрын
@@davidgilardi6923 Great! Would you mind sharing the workflow with us?
@davidgilardi6923 12 күн бұрын
@@TheHallofFameVault KZbin won't let me post links. However, if you'd like feel free to reach out on LinkedIn at david-gilardi or email me at [email protected]. I'm happy to share the flow. Better yet, can also find the example I used in Langflow as a template. Just choose +New Flow -> Agents (very bottom of the list) -> Market Research.
@TheHallofFameVault 11 күн бұрын
@@davidgilardi6923 Please check your email. I have sent messages to you. Thanks
@marcosoliveira8731 19 күн бұрын
Really nice video. +1
@DataStaxDevs 18 күн бұрын
Thanks for tuning in!
@sriniBw 19 күн бұрын
Sir, I have tried using the Ollama model and need to connect it with the agent to access the tools. However, I am unable to do so. Is there any possible way to connect Ollama with the agent? Thank you.
@DataStaxDevs 19 күн бұрын
We're investigating and will get back to you!
@davidgilardi6923 18 күн бұрын
Hey there sriniBw. Ok, you can definitely connect an Ollama model to an agent in Langflow. I just went through some scenarios. In the Agent component, choose "custom" for the Language Model dropdown. Then, pull over your Ollama component and wire up the purple "Language Model" node to the purple "Language Model" node on the agent. Here's a list of models that should work with tool calling in Ollama. ollama.com/search?c=tools However, I just used qwen2.5:latest and it is working nicely.
@davidgilardi6923 18 күн бұрын
I'm happy to make a video on this topic.
@sriniBw 19 күн бұрын
Sir, I have tried using the Ollama model and need to connect it with the agent to access the tools. However, I am unable to do so. Is there any possible way to connect Ollama with the agent? Thank you.
@DataStaxDevs 18 күн бұрын
We are investigating!
@creapics 22 күн бұрын
great video, thanks a lot
@DataStaxDevs 21 күн бұрын
glad it was helpful!
@SelcukMustafa-n1r 22 күн бұрын
This could really change the way we approach job hunting. I’ve heard a bit about Resubot, and it’s interesting to see how AI can help in this space.
@CemileZekeriya 22 күн бұрын
Very informative session. I’ve been using Google Docs for my resume, but I’m curious if tools like ResuBot would save more time.
@JannieDunham 22 күн бұрын
The integration of GenAI into the job search is fascinating. Looking forward to trying out Astra DB and Langflow.
@DataStaxDevs 22 күн бұрын
Thanks for tuning in!
@TERRENCEMoriarty 22 күн бұрын
Prompt engineering sounds complex but essential. I hope to see more tutorials on using these tools effectively.
@DataStaxDevs 22 күн бұрын
Thanks for tuning in!
@BARNEYLlamas 22 күн бұрын
This is a great approach to streamline the job application process. I wonder how Resubot compares to this method for creating resumes.
@hardinrami8500 23 күн бұрын
you are the best!
@ArekMateusiak 25 күн бұрын
Hi, great work! It would be cool to have more types, and dropdown list for types. Also it would be great to have required checkbox, and for example enum type list to select from for model... also limits for characters also could be awesome to control how many characters we permit model to produce.
@davidgilardi6923 24 күн бұрын
Yes, I totally agree. I'll pass your feedback along to the team. Thanks for your thoughts! :D
@AchilleasDrakou 26 күн бұрын
Tejas you are the goat. Great tutorial, very information dense!
@mehdihachana_3138 Ай бұрын
this was really helpful thanks a lot dude
@DataStaxDevs 29 күн бұрын
thanks for watching!
@roushansingh8993 Ай бұрын
does embedding astra db with langflow, and get output for free? I am asking because, I am getting errors like Error code: 429 - {'error': {'message': 'You exceeded your current quota, please check your plan and billing details
@DataStaxDevs 26 күн бұрын
Hi there! Third party embeddings services such as OpenAI are not free with Astra/Langflow. The 429 error is likely related to your OpenAI account. We recommend checking your OpenAI credit balance and make sure it is positive: platform.openai.com/settings/organization/billing/overview Or you can swap out OpenAI for another embeddings provider (HuggingFace and Astra Vectorize with NVIDIA are the two free options, for example)
@NIRMALTHULASINGAM Ай бұрын
Hi I'm having an error while importing CassIO. The error arrises like depricated exception for 'import cassio'. It displays to install C++ build tools which is already installed.
@DushyanthG Ай бұрын
Thank you Tejas; For the practical explanation!
@DataStaxDevs Ай бұрын
🙌 glad you found it useful!
@fahrankamili7931 Ай бұрын
this is such a good talk. 4k views is criminally underrating this video
@DataStaxDevs Ай бұрын
🙌 glad you found it useful!
@banihas22 Ай бұрын
Is there a JSON file for Langflow or the code from the app portion you can share from this talk?
@DataStaxDevs Ай бұрын
We will ask David and circle back!
@AbhishekGautam-Y1 Ай бұрын
How the hallucinations were avoided?
@stebashley Ай бұрын
Great explanation!
@DataStaxDevs Ай бұрын
Glad you found it useful!
@ammarparmr Ай бұрын
Thanks but why there is no embedding for the PDF file and just parse the text?
@ammarparmr Ай бұрын
Short & Precise! Thanks
@DataStaxDevs 29 күн бұрын
thanks for tuning in!
@stefano94103 Ай бұрын
Unhhelpful
@hyperborean72 Ай бұрын
after following this tutorial for the first 40 minutes the first questions that came to my mind are: 1) how partitions are related to each other in Cassandra. I mean how in a situation without master data are sent to this or that 'ring' based on their partition key 2) how AstraDB is related to Cassandra. Is AstraDB some tool for easy Cassandra cluster deployment also providing API for different programming languages, or is it the separate database 3) how Cassandra/AstraDB is dependent on cloud providers (Google, Amazon, Microsoft). Do I understand right that cloud providers provide dataspace for the Cassandra nodes. If so, why do they do that. ... I admire how clean and detailed the explanation, thank you.. if only you placed emphasis always correctly
@JeremyDevz Ай бұрын
Getting into RAG and this was great. Awesome and informative video, Sonic! 👍🏽
@DataStaxDevs Ай бұрын
That's great to hear!!
@romankhairov Ай бұрын
its strange that with a minor update the application changed many basic nodes
@ahmetyusuf9964 2 ай бұрын
code ?
@TopMenMotivation 2 ай бұрын
Interesting thanks.
@ShrirangKanade 2 ай бұрын
is it how we implement a message queue? how can we clear this queue after i have done my messaging? it would be helpful if you answer.
@rajkrupz 2 ай бұрын
Thank you, may I please know how to get the exercise you mentioned at the very end?
@bhendegaa 2 ай бұрын
Here, billions of partiontions not causing issue, I did not get that point as partition key has limitation in terms of data modeling best practices where wide portions should be 100MB, other wise excessively wide partitions have a negative impact on performance and are not recommended. A partition is considered to be wide when the size is greater than 100 MB. The practical limit on the size of a partition is two billion cells, but it is not ideal to have such large partitions. The maximum partition size in Cassandra should be under 100MB and ideally less than 10MB. Application workload and its schema design haves an effect on the optimal partition value. However, a maximum of 100MB is a rule of thumb. A ‘large/wide partition’ is hence defined in the context of the standard mean and maximum values.
@spaceninja5709 2 ай бұрын
What is this and how can i learn it
@DataStaxDevs 2 ай бұрын
check out Langflow.org to learn more!
@daniel_bartosiewicz 2 ай бұрын
How do I connect Langflow to external application? I tried, it doesn't seem to work.
@merdandt 2 ай бұрын
Amaizing! Waiting for more tutorial on different use cases from you guys!
@DataStaxDevs 2 ай бұрын
Great to hear! Lots more to come!
@dix-neufsoixante961 2 ай бұрын
She's is definitely not a developer lol