DeepSeek-R1: Open Source O1 Rival in 6 Minutes

--- type: transcript date: 2025-01-20 youtube_id: Or0ruWJybVI --- # Transcript: DeepSeek-R1: Open Source O1 Rival in 6 Minutes a huge release out of deep seek today so deep seek R1 their reasoning model is now fully open source with an MIT license once interesting with this model is it is the first major competitor to open AI 01 series of models in this video I'll give you a brief overview about the model the pricing some of the technical details and then I'll show you how to get started with the model right off the bat to go over some aspects about this model it's a mixture of experts model in terms of the active parameters it's 37 billion active parameters what's interesting with this is it's actually the same size as deep seek V3 for the mlu score it scores a 90.8 in comparison claw 3.5 Sonet has 88.3 GPD 40 has 87.2 respectively and this even outperforms 01 mini and it is just shy of open AI 01 that was just released last month just about all of the metrics for the coding benchmarks with the exception of ader polyot are just shy of 01 but where you can really see the difference on some of these benchmarks is look at code forces for instance Claude 3.5 Sonet scores 20.3 gbd4 23.6 R1 scores 96.3 another huge thing with this release is it has a very permissive license it's MIT license you can distill the model you can commercialize the model you can use the outputs to fine-tune other models or use the outputs from the model for synthetic data generation or basically whatever you want they also released six small models that are fully open source as well these are open source distill models from R1 we can see here deep seek R1 we have 1.5 70b 14b all the way through to 70b and the 32b and 70b these are on par with opening eyes 01 mini what's great with these distilled models is you're going to be able to run this basically regardless of the hardware that you have now to quickly touch on the technical report I'll also link everything within the description of the video if you're interested we introduced our first generation reasoning models deep seek r10 and deeps R1 deeps r10 a model trained via large scale reinforcement learning without a supervised fine tuning as a preliminary step demonstrating remarkable reasoning capabilities through RL deep seek r10 naturally emerges with numerous powerful and intriguing reasoning behaviors however it encounters challenges such as poor readability and language mixing to address these issues and further enhance reasoning performance we introduce deep seek R1 which incorporates multi-stage training and cold Data before RL deep seek R1 achieves performance comparable to opening eyes 01217 on reasoning TS to support the research Community we open source deep seek r10 deep seek R1 1.5b 7B 8B 14b 32b 70b which were all distilled from Deep seek R1 based on quen and llama if you're interested in diving in a little further I'll link everything with in this technical report within the description of the video to try out the model you can try it out at chat. Deep seek. what's really great with this is you also see the reasoning steps here which is different than what you get from open ai1 where you don't actually see here we see our short story but now if we actually test some of the reasoning capability if I say how many times did the letter R occur in that story I'll put that through and we see okay the user wants to know how many times the letter R appears in the story what it's doing is it's breaking up all of the sentences and then with each sentence it's breaking up every single word and it's counting word by word at the bottom we have the breakdown of all of the different paragraphs and potentially how many times the letter are occurred within that now in addition to the chat interface you can also access this from their API with the model string of deep seek Reasoner next what's really unreal with this model is the pricing for a million tokens per input it is 55 cents with context caching is 14 and then for the output tokens it is $219 so if we compare that to 01 output tokens are $60 a million tokens of input is $15 and context caching is still $7.50 per million tokens the pricing alone from their hosted API makes this model incredibly interesting in terms of some of the specifics of their API defaults to 4,000 tokens of output with a maximum of 8,000 tokens but with a Chain of Thought the output can be as high as 32,000 tokens there isn't the ability to control the reasoning effort right now like you can with the 01 models what I love with the output is you can get both the reasoning content so all of those steps and then parsed out individually in a separate key is the actual content of the response as well aravan shavas mentioned that deep seek has largely replicated 01 mini and has open sourced it from technium put it well destroys 40 we have Frontier models at home I guess now another thing worth mentioning is it's free to use on the web page what I demonstrated for you with the exception of the API that is all free and you can also download a mobile app that they just released the last I just want to show you how you can set this up within your IDE if you're interested so continue is a tool that I love you can use this within vs code wind surf or cursor and what you'll be able to do is once you've installed continue from the extension Marketplace you can go to the bottom here and add in a new chat model what you can do here is even if you don't see the reasoning model from the dropdown what you can do is within the configuration file is if you just set the model string to deep seek Reasoner you set the context length and you put in your API key you'll be able to use this within your IDE if I just demonstr here if I just say write me a proxy server for opening eyes chat completion endpoint pretty quickly we see the responses here it's streaming all back and you can use it similar to something like you would cursor I can just add that all in here and we have what looks to be like a great response so this is a really good option if you're looking to use this in the context of an ibe and the thing that I noticed with this model especially from their API is what is very very quick overall that's pretty much it for this video kudos to the team over at Deep seek on this release it is really great to be able to have these types of models out there within the open source Community let me know your thoughts in the comments below and if you found this video useful please like comment share and subscribe otherwise until the next one