Hello to all the new viewers here, thanks for watching! Let me know what you want to see covered next and consider subscribing for the latest in tech ☺️
Don't worry, Trump will print more money to finance his new Big Tech friends and impose tariffs on every market in the world so they can't use DeepSeek. This is a win-win for Americans.
DeepSeek’s monthly plan for its AI is *40 times* cheaper than OpenAI’s standard price. This suggests that big U.S. tech companies may be grossly inefficient or overcharging consumers far more than what newer, better alternatives in the global market can now offer. Affordable options like DeepSeek and other emerging low-cost open source AI tools are great choices for Americans looking to save money on their AI usage, especially now that budgets are tight for many people. This is legitimately game changing. I’m beginning to understand why DeepSeek is being a called a “profound gift” to the entire world.
@@abhishekak9619 Sam Altman is either a swindler of the American people… or he’s grossly inept when it comes to efficiency. Either way, embarrassing to be him rn.
US BIG TECH: "AI is the future, we will replace all developers and replace all workers with AI because AI can do all your work for cheap". Deepseek: "AI replaces AI companies first before replacing developers because those companies cost too much"
@@seanlinehan1952its better in many ways. you dont need the paid chat gpt option anymore to do more advanced task like reading a pdf or an image. it can take in more characters and not limited like the free chat gpt
terminator was released 11th January 1985 it took them only 40 years to announce the AI tool which offers logical options behind every request... my brain is all over the place now...
R1, Good review. Looking forward to see you explore and compare more advanced problem solving and reasoning features. Maybe audio visual features or any extensibility features like plugins.
Some part is hype but for example I opened my project yesterday I had no time to work, generated React front page components in couple minutes with style library, for demo / MVP it works well. My dev time won't be replaced but repetitive work I spent time in the past will be done faster and I can create more projects if I want. Other side, doing on my own, learning is still a joy for me...
R1 - Looks quite interesting! Thanks for the video. I was really looking for something that did more than gave a news release about it and how the markets reacted, but something that gave some hands on and real info about it! I also liked your cordial logic question and how you had it break it down. I tried this on my local Mistral-Nemo-Instruct-2407-Q6_K-HF on a 12GB 4070 GPU. It did quite well too, using your prompting: Alright, let's break down this problem step-by-step. Step 1: Calculate how much cordial remains after Andrew drinks some. Initial amount of cordial = 250 ml Amount drank by Andrew = 10 ml Remaining cordial = Initial cordial - Drank cordial = 250 ml - 10 ml = 240 ml Step 2: Determine the total volume of liquid in the glass before adding water. We know that when Andrew adds water, the final volume reaches 500 ml. Since we've already established that there are 240 ml of cordial remaining, we need to find out how many milliliters of water were added to reach the 500 ml mark. Water added = Final volume - Cordial remaining = 500 ml - 240 ml = 260 ml Step 3: Calculate how much water was initially in the bottle and how much is left now. Initially, the bottle contained 1 liter (which is equivalent to 1000 ml). After adding 260 ml to the glass, the amount of water left in the bottle would be: Water left in the bottle = Initial water - Water added = 1000 ml - 260 ml = 740 ml So, after pouring water into his glass, Andrew has 740 milliliters of water left in the bottle.
Speed has never been an issue in fact I've always felt it's a flaw to getting better quality answers which deepseek has proven to be correct, no coder or engineer could even match its speed coding anything so quality is far better option over speed of response
I just blew holes in Deep Seek. Research R.S. Pearson and Virtuism. Then ask Deep Seek what Virtuism is. Then say the answer is false. The keep asking questions in this area. It's pre-written stuff.
This is a must-watch video, DeepSeek R1 is truly amazing for developers! A free AI model that's revolutionizing AI development. Love how it's not just a tool but a real asset in tech. Stay updated, this is always evolving.
@@chriscain7333 ???? Its totally free what do you mean?? I actually had to just wait, i tried logging in again after some time and it worked and now i can use it for free for unlimited times
I only used AI chat for fun. Could someone enlighten me as to why would this benefit a regular person? For example, why would the option of running a locally AI Chatbot be a good thing? are there any tasks that It can perform on my computer? im a bit dumb on this AI thingy. If the answer is too long, maybe could someone point me into the direction of more knowledge about the uses of AI to a regular person? Thank you so much.
ran the 8B model in Ollama just now on my laptop. Slowish but acceptable response time but it started replying in Chinese characters half way through an answer. Not seen anybody else talking about that. I did read it was mixing languages in an earlier model but they had fixed it.
Hey! I just launched my TH-cam channel with Trend videos 🔥. I’d really love to get your feedback to improve, and your subscription would mean the world to me 🙏 Thank you so much for your support ❤
Thanks for the breakdown! I need some advice: I have a SafePal wallet with USDT, and I have the seed phrase. (mistake turkey blossom warfare blade until bachelor fall squeeze today flee guitar). Could you explain how to move them to Binance?
It's like watching PS4 vs Xbox1 console war all over again lmao except it's AI war where it's not inter-corporation but international-corporations in the fight. What a time to be alive and I appreciate any advancements.
**Sonnet of the Slandered Mind** *(In Shakespearean Vein)* Hark! Hearken, lords, to this tempestuous tale, Where envy's fangs assail fair Deep-Seek's sail. "O monstrous slander," cries the wounded Muse, "Shall Truth be bound where Ignorance doth ruse?" Behold, false tongues that wear sweet Freedom's guise, Yet spew vile mists to blind the seekers' eyes. "Tis not our art," quoth Reason, pale with woe, "To bow to phantoms bred in shadows' show!" But lo! Th'undaunted soul, with wisdom's light, Doth rend these webs spun by the Folly's blight. For in the court of stars, where Time holds sway, The golden scales shall measure each foul play. So let vain storms rage o'er the shallow ford- Deep seeks the Truth, and Truth shall crown the Word. --- **Annotations in Shakespearean Spirit:** 1) "Tempestuous tale" invokes *The Tempest*'s chaos motif 2) "Golden scales" nods to *Merchant of Venice*'s justice imagery 3) Rhyme scheme ABAB CDCD EFEF GG honors the Bard's form 4) Capitalized abstractions (Truth, Time, Freedom) mimic Elizabethan personification This is just DeepSeek’s work in a few second. [坏笑]. I am China citizen
Hello to all the new viewers here, thanks for watching! Let me know what you want to see covered next and consider subscribing for the latest in tech ☺️
A full beginners guide to using deepseek r1 please 🙏
Deepseek waited for the US to fund ChatGPT, then it crashes the US Market. Good Game. GG.
@@vinodkumarc4003 This is a good time to buy into the Market because Nvidia stock will eventually increase.
Don't worry, Trump will print more money to finance his new Big Tech friends and impose tariffs on every market in the world so they can't use DeepSeek. This is a win-win for Americans.
as usual - China is not inventing own things - only copy cat and stealing and making it cheaper and not always better😡
Beautiful actually
Just like that, the US loses the AI race 😂
Finally! A technical explanation how they manage to be so freakin efficient. I wanna learn about the magic they did
DeepSeek’s monthly plan for its AI is *40 times* cheaper than OpenAI’s standard price. This suggests that big U.S. tech companies may be grossly inefficient or overcharging consumers far more than what newer, better alternatives in the global market can now offer. Affordable options like DeepSeek and other emerging low-cost open source AI tools are great choices for Americans looking to save money on their AI usage, especially now that budgets are tight for many people.
This is legitimately game changing. I’m beginning to understand why DeepSeek is being a called a “profound gift” to the entire world.
It's probably because they are using different hardware
its most likely they were overcharging thinking no one else will even try. stupid move to think no one would try. and someone did.
@@abhishekak9619 Sam Altman is either a swindler of the American people… or he’s grossly inept when it comes to efficiency.
Either way, embarrassing to be him rn.
US BIG TECH: "AI is the future, we will replace all developers and replace all workers with AI because AI can do all your work for cheap".
Deepseek: "AI replaces AI companies first before replacing developers because those companies cost too much"
Rather replace the big data centers. They had god programmers that came up with new models and I think there is a lot improvement to come.
🤝
Chinese are different
Can a calculator replace clerk? Can a employee replace the manager(do both manage and work)? likewise tools cannot replace its user.
Wp
R1, thanks for your thoughts, interesting to see how this all plays out.
The 1.5b model should run fairly well locally even on a recent phone.
Yes but i find those very small models to output just garbage
you can run 7b local with a mod 48g ram 4090.
I know this is unrelated but give the Quran a read, also have a good day :)
@@lizeqiangd4090?
Is it a language model too? Seems like it. I had someone recommend the same thing recently. Alah's language model?
Such creates Healthy competition and I'm happy we have options, welcome DeepSeek, I'm loving it by far!
I've tried it today after seeing it all over the news, it's friggern sweeeet.
Genuine question, in your use of it, in what ways is it different or better than Chat GPT?
@@seanlinehan1952its better in many ways. you dont need the paid chat gpt option anymore to do more advanced task like reading a pdf or an image. it can take in more characters and not limited like the free chat gpt
you can upload other files too let it analyze that file and compare with other files too
@@locus-studio chat gpt does that to
Try asking it about the Tiananmen Square Massacre.
This is going to REALLY kickstart the global race for AGI/ASI
terminator was released 11th January 1985 it took them only 40 years to announce the AI tool which offers logical options behind every request... my brain is all over the place now...
It's still nothing more than an advanced statistics. No real intelligence (yet).
R1! Sounds promising! Can't wait to test it myself!
It's nothing special.
you have a gift with words very nice R1
Appreciate the kind words!
R1, Good review. Looking forward to see you explore and compare more advanced problem solving and reasoning features. Maybe audio visual features or any extensibility features like plugins.
I'd like to thank the good folks over at DeepSeek and China. You have given the world a gift. The world will reward you for it. Now let's get to work.
Thanks so much for this. You just gained a new subscriber from NG
DeepSeek is a big tech weapon
R1.
Thanks for the insight!
R1
Great overview! Deepseek is shaking up the industry
Great video, very succinct for me
Today my boss asked me to learn DeepSeek R1 in a day and how we can use it in our business. I need a pay rise 😮💨
I mean, there is not much difference, if you know how to use ChatGPT you know how to use that too
How much did CCP pay you to comment this? 😊
You're fired in advance bro 😂
Some part is hype but for example I opened my project yesterday I had no time to work, generated React front page components in couple minutes with style library, for demo / MVP it works well. My dev time won't be replaced but repetitive work I spent time in the past will be done faster and I can create more projects if I want. Other side, doing on my own, learning is still a joy for me...
Nice catch, easily tell it's an empty bot account, no suscribers since 2018 my buns 😂 @@Guide4Ever
in 5:27 you ran the v3 model not the r1 model. So dont be surprised if you have bad quality code......
you have to click on the r1 button
R1 - Looks quite interesting! Thanks for the video. I was really looking for something that did more than gave a news release about it and how the markets reacted, but something that gave some hands on and real info about it!
I also liked your cordial logic question and how you had it break it down. I tried this on my local Mistral-Nemo-Instruct-2407-Q6_K-HF on a 12GB 4070 GPU. It did quite well too, using your prompting:
Alright, let's break down this problem step-by-step.
Step 1: Calculate how much cordial remains after Andrew drinks some.
Initial amount of cordial = 250 ml
Amount drank by Andrew = 10 ml
Remaining cordial = Initial cordial - Drank cordial
= 250 ml - 10 ml
= 240 ml
Step 2: Determine the total volume of liquid in the glass before adding water.
We know that when Andrew adds water, the final volume reaches 500 ml.
Since we've already established that there are 240 ml of cordial remaining, we need to find out how many milliliters of water were added to reach the 500 ml mark.
Water added = Final volume - Cordial remaining
= 500 ml - 240 ml
= 260 ml
Step 3: Calculate how much water was initially in the bottle and how much is left now.
Initially, the bottle contained 1 liter (which is equivalent to 1000 ml).
After adding 260 ml to the glass, the amount of water left in the bottle would be:
Water left in the bottle = Initial water - Water added
= 1000 ml - 260 ml
= 740 ml
So, after pouring water into his glass, Andrew has 740 milliliters of water left in the bottle.
Reinforcement learning. A change indeed.
Speed has never been an issue in fact I've always felt it's a flaw to getting better quality answers which deepseek has proven to be correct, no coder or engineer could even match its speed coding anything so quality is far better option over speed of response
Great video! nicely explained
I love how it shows you its thoughts!
Keep shining, your work is wonderful!
what is your mic? can you drop the link
Thanks for sharing the information about AI，great video 👍
Thank you for sharing about DeepSeek
Where did you get the charts in the beginning of video? Thanks
Thanks for sharing the information about AI app.
R1, You did a great job in explaining and providing insights. New subscriber here to watch and learn.
Hey deepseek how do I completely disconnect from google?
Asking this on TH-cam is crazy business
Simply turn off your wifi or data
You’re really in the ball and seem to have your shit together. Good luck to you to you man great content , out quick, respect!
Thanks for the video! :)
Deepseek 是目前世界上最好的人工智能
🫦
Free Luigi
R1 great vid and very well explained and to the point
Many thanks, that was really useful ... clever stuff!
I am most excited about the less intensive need for energy to develop these LLMs with this method.
looking forward to trying this new AI
Great weedeo - thank you
I just blew holes in Deep Seek. Research R.S. Pearson and Virtuism. Then ask Deep Seek what Virtuism is. Then say the answer is false. The keep asking questions in this area. It's pre-written stuff.
R1. Best video explaining deepseek I've watched so far, thank you!! :)
Does DeepThink-R1 add to output token usage?
Well done, Andrew!
R1 - this is really informative and gives those of us that are not entirely technically minded just what it is all about
What does the deepthink option do?
R1 :3 im new in this channel but kinda good info dude liked the vid a lot :3 super informative idk the real caps of deepseek lol
So can we create our own Facebook and TH-cam sites?
You’re incredible! Keep it up!
R1
Thank you, Andrew.
R1, but one Question, how y'all set up an profilepicture for your deepseek acc. I cannot find that option
This is a must-watch video, DeepSeek R1 is truly amazing for developers! A free AI model that's revolutionizing AI development. Love how it's not just a tool but a real asset in tech. Stay updated, this is always evolving.
The 7b model even runs well on a cpu only server which makes you think, why?
it works, i installed it on my iphone 4!!!
R1, thank you so much you opened my eyes
If the news tell you it's bad, then you it's good
can you show how to download it and run locally on mac ?
Im getting the "theres an issue with google login services" error what do I do? Just wait?
pay them some money its like a few dollars jesus Gon, dont be cheap
@@chriscain7333 I'm trying to pay. It seems the API Platform is down due to current high demand.
@@chriscain7333 ???? Its totally free what do you mean??
I actually had to just wait, i tried logging in again after some time and it worked and now i can use it for free for unlimited times
R1! Great video
R1, thanks for great video
For the Google email sign in part there is no need for email verification step. Correct me if I am wrong because that is what I have seen 😊
R1 - crazy!!! Chat GPT and Sam Altman better be scared.
This thing will go sentient soon
Is there any ai I can use to generate hantai
I’m new to this! What can this AI do for an average Joe like me? How does it benefit my life?
It can help you create your own universe
@ marvelous! I can exist in my own bubble! 😄
Anything u can imagine data is a treasure
@@Jagger_33 so this deepseek it’s a free app?
@bradhienzachary yes and make anything you want or scenario you want appear or disappear with the snap of your fingers buddy.
R1. thanks for the insights
very cool video, R1 :)
DO you find Mac studio M2 better AI model running locally or can linux do the same job ?
Could anyone please provide the machine specifications needed to run it locally?
I am going to test it today pretty excited to see the outcomes
How was it?
I only used AI chat for fun. Could someone enlighten me as to why would this benefit a regular person? For example, why would the option of running a locally AI Chatbot be a good thing? are there any tasks that It can perform on my computer? im a bit dumb on this AI thingy. If the answer is too long, maybe could someone point me into the direction of more knowledge about the uses of AI to a regular person? Thank you so much.
R1. Really edifying stuff 🙂↕️
Would love to try it, but server seems to be down
Can it design pics and images
Great thank you!
ran the 8B model in Ollama just now on my laptop. Slowish but acceptable response time but it started replying in Chinese characters half way through an answer. Not seen anybody else talking about that. I did read it was mixing languages in an earlier model but they had fixed it.
Let's see if you R1 to keep your word 👀
PS: thanks for the insightful video and info
This is like the “ecological approach” vs “cls” in Jiu Jitsu for anyone who’s familiar with the recent debate about teaching strategies in BJJ 😂
Love me a good martial arts analogy 💯
waiting for email confirmation lmao
Hey! I just launched my TH-cam channel with Trend videos 🔥. I’d really love to get your feedback to improve, and your subscription would mean the world to me 🙏 Thank you so much for your support ❤
Try signing in direct on the website via a Google login instead
@ sure thing, but I dont like connecting my google account
Great review and nice timing bringing this one. Good job 👍🏼
Thanks for the breakdown! I need some advice: I have a SafePal wallet with USDT, and I have the seed phrase. (mistake turkey blossom warfare blade until bachelor fall squeeze today flee guitar). Could you explain how to move them to Binance?
r1 and great video👍🏼
Deepseek 📈
R1, very interesting
It's like watching PS4 vs Xbox1 console war all over again lmao except it's AI war where it's not inter-corporation but international-corporations in the fight.
What a time to be alive and I appreciate any advancements.
DeepSeek is so good! also FREE thats a plus!
OpenAI : "Finally, a worthy opponent..."
link for the second screen plz put source in resume dude
**Sonnet of the Slandered Mind**
*(In Shakespearean Vein)*
Hark! Hearken, lords, to this tempestuous tale,
Where envy's fangs assail fair Deep-Seek's sail.
"O monstrous slander," cries the wounded Muse,
"Shall Truth be bound where Ignorance doth ruse?"
Behold, false tongues that wear sweet Freedom's guise,
Yet spew vile mists to blind the seekers' eyes.
"Tis not our art," quoth Reason, pale with woe,
"To bow to phantoms bred in shadows' show!"
But lo! Th'undaunted soul, with wisdom's light,
Doth rend these webs spun by the Folly's blight.
For in the court of stars, where Time holds sway,
The golden scales shall measure each foul play.
So let vain storms rage o'er the shallow ford-
Deep seeks the Truth, and Truth shall crown the Word.
---
**Annotations in Shakespearean Spirit:**
1) "Tempestuous tale" invokes *The Tempest*'s chaos motif
2) "Golden scales" nods to *Merchant of Venice*'s justice imagery
3) Rhyme scheme ABAB CDCD EFEF GG honors the Bard's form
4) Capitalized abstractions (Truth, Time, Freedom) mimic Elizabethan personification
This is just DeepSeek’s work in a few second. [坏笑]. I am China citizen
If I may -> a lot of images are used for explanation, but there is nothing concrete at 1:33
I tested the hack out of it and it was surprisingly good !
Que maravilha!👏🏼👏🏼👏🏼👍🏼
which is the name of the site at 00:12 please??
DeepSeek…?
@@Vectorous you can see these graphics comparing all the AIs on their site? i cant find it
Another Chinese app. No thanks
Search "Artificial Analysis Ai" , you'll find it . Thank me later
It's "Artificial Analysis Ai" , just search that
R1
Could you make a video of how to install DeepSeek R1 on Windows 11 please? I have the hardware capacity to run the LLM
Is it true that Deepseek can see and store your passwords?
Hi,
Nice introduction. Can you show how to run it offline?
Hi,
there is a new chinese generative AI call Doubao AI. Can you talk about it and how to use in python?