Відео

what does this have to with the saturn cube

my project suddenly lost all the functions including the match_documents function, thanks to your 5:50s solution I was able to solved it. I was 4 hours trying to solve it, thank you man you made my night

Whoever talks about Deepseek is a traitor who supports Chinese Communists. I will unsubscribe now. You are giving all your private information to China.

You guys are traitors sponsoring Communists. Blocking you.Have some dignity. If you cant afford burger cost why you need AI, stupid

Deepseek sold my baby.... #meatbag

For planning, as a basic chain before the main orquestrator, is best to use the 70b llama distilled one from groq because its way faster and you get 80% of the benefits from the thinking even at 70b

I have a starter subscription on n8n. I do not see community nodes in settings. Is it not available for starter?

Hey thanks for this video, did you know if we can change the provider for each model (R1 and V3) because the default one is x3 expensive than the basic one but i don't know how we can change this in n8n. Thanks !

exiting!

Thanks ! Would be interesting of a DeepSeek R1 API version that can help to switch fast from OpenAI API.

I dont see the Base URL in chatgpt..

Hi, how can I create the Jarvis voice in elevenlabs ?

Probably one of the best I have seen. Intuitive easy to follow

IDK how to say it, but you saved me so much time and stress. its working now like a charm. so 1000 thanks. god bless ya.

This was straight-to-the-point and helpful. Thank you for the quality content.

Non-reasoning model, I sleep.

great vid

Wait, can I create AI agents with this basically?

"don't download the 6-70-parameter model, it's going to destroy it" like aluminum in the microwave

chaos, disruption, the more the better! no cooperation or love

I'm going to drop a golden nugget in here, coming from real world experience and many agent executions, not theory. Best match made in heaven for me has been: Qwen 2.5 instruct 72b (which im planning to upgrade to MAX when it becomes available in openrouter) as an agent orquestrator (executor) and having Deep Seek R1 as a tool of the first agent. Then controling its behaviour via a well crafted system prompt for each one of them with enough dynamic context. This is making Qwen call R1 when its coming short of thinking and returning a 'suggest plan of action' from R1, pretty much outsourcing the thinking and coming back with a plan of action that Qwen executes flawlessly elevating the output's quality by a lot.

Its beating DeepSeek? You noticed that the benchmarks are compared to the V3 model, not the R1, right? Please compare the performance of 2.5 Max to R1 to actually be able to tell whos beating who.

out of the topic- kindly make also a tutorial in n8n using gemini flash 2.0 for booking and reservation.

Thanks !!!

Link in the description?

Another one 😮

yo

How can we use this on VPS?

I m learning more from the thinking process than the actual answer prompts. Get out chatgpt openAI grok meta. Soon all pc laptops will come preinstalled deepseek locally.

Nerd stuff..... not suited for the average user.

If i want to run this in the cloud because my computer is too old, what will you recommend at a good price.

hello i have this error "'Problem in node ‘AI Agent" does someone know how to fix this ?^^ please help

Merci beaucoup. Simple et efficace.

ARE THEY STORING IT IN CHINA , WITH ALL YOUR INFORMATION!!!???😅😊

Try misspell the word strawberry as strawberrry and ask the "how many Rs" question again . The thinking process is even more impressive.

How to make it stop thinking and repeating the input

How do you handle automation with n8n for client projects while ensuring data security after completion? For example, do you work with a prepaid server, a new Google account, and a fresh OpenAI API key, allowing the client to change credentials afterward? Or do you sign a non-disclosure agreement and manage everything on your own account and server? I'd love to hear how you approach this!

Bro this is HILARIOUS! I absolutely love it 😂😂

thanks man, another nice video. however: - not sure DeepSeek supports function calls hence Tools Usage via AI Agent Node might NOT work - best use case scenario for DeepSeek model with reasoning capabilities would probably be agent orchestration, so that the actual Tools Usage can be delegated to OpenAI Chat Model (that surely works) as opposed to Ollama Chat Model (that as of today does not work with Tool Usage)

TY! Well explained

📚 Join my Skool community (access all n8n templates & more): www.skool.com/aiworkshop 🤝 Need Al solutions built for your business? Work with me: aiworkshop.me/contact/

Does this agent have memory? can you ask follow-up questions ?

I'm building a custom, dedicated pc for AI/LLM with multiple SSD drives. Is there a way when downloading the models, to choose a specific drive to download the model to? Or will the download only be to the drive with the OS installed?

DeepSeek❤️ China💥

Does it work with tools? Grow implemented it for free but it seems it doesn't support tool agents when you add any tool in n8n

Thanks man🫡

do you have a tutorial on how to install custom npm packages to use in n8n?

do you have a tutorial on how to install custom npm packages to use in n8n?

Connecting it to Elevenlabs can become super expensive…