1 00:00:00,000 --> 00:00:00,000 Guys. 2 00:00:00,000 --> 00:00:02,000 Here is an amazing crash course on Amazon. 3 00:00:02,000 --> 00:00:03,000 Bedrock. 4 00:00:03,000 --> 00:00:05,000 Uh, many people were requesting for this. 5 00:00:05,000 --> 00:00:08,000 Uh, now, what exactly is Amazon bedrock? 6 00:00:08,000 --> 00:00:14,000 It is the easiest way to build and scale generative AI applications within your AWS platform itself, 7 00:00:14,000 --> 00:00:15,000 right? 8 00:00:15,000 --> 00:00:16,000 Right. 9 00:00:16,000 --> 00:00:18,000 Now many companies are using this. 10 00:00:18,000 --> 00:00:23,000 They are using this because understand one thing that there are so many different, different companies 11 00:00:23,000 --> 00:00:29,000 who are providing LLM models for or LLM models for different different text generation and image generation 12 00:00:29,000 --> 00:00:29,000 task. 13 00:00:29,000 --> 00:00:32,000 Right there is open eye, there is cloudy two. 14 00:00:32,000 --> 00:00:35,000 There is Google, there is even. 15 00:00:35,000 --> 00:00:39,000 Amazon also has its own LM model which is named as Titan. 16 00:00:39,000 --> 00:00:42,000 And similarly there is also Facebook which is like llama two model. 17 00:00:42,000 --> 00:00:45,000 So understand each and every thing is that. 18 00:00:45,000 --> 00:00:49,000 What is the main problem right now if you want to use these APIs, if you want to use these models, 19 00:00:49,000 --> 00:00:53,000 everybody has a different, different setup right now. 20 00:00:53,000 --> 00:00:55,000 What Amazon Bedrock actually provides you. 21 00:00:55,000 --> 00:01:02,000 It provides you one AWS platform where all the models will be available, and through their API calls. 22 00:01:02,000 --> 00:01:04,000 You can probably use any of this kind of models. 23 00:01:04,000 --> 00:01:09,000 Right now, OpenAI is not there, but other than that, I could see almost each and every model that 24 00:01:09,000 --> 00:01:11,000 was available in Amazon bedrock. 25 00:01:12,000 --> 00:01:18,000 Now, why this is important because a single platform here, you don't have to worry about scalability. 26 00:01:18,000 --> 00:01:20,000 You don't have to worry about other things. 27 00:01:21,000 --> 00:01:25,000 Uh, with respect to the cost, that is, uh, uh, in the Amazon bedrock, right? 28 00:01:25,000 --> 00:01:28,000 It is slightly more than the OpenAI. 29 00:01:28,000 --> 00:01:33,000 And probably in the future it may also get reduced, but at the end of the day, you should know what 30 00:01:33,000 --> 00:01:34,000 exactly Amazon Bedrock is. 31 00:01:34,000 --> 00:01:39,000 And probably I'll show you in this particular video different, different tasks. 32 00:01:39,000 --> 00:01:42,000 I will try to code it, and I'll try to show you how you can use these APIs itself. 33 00:01:42,000 --> 00:01:47,000 So here is what is Amazon Bedrock and easiest way to build and scale generative AI application with 34 00:01:47,000 --> 00:01:48,000 Foundation model. 35 00:01:48,000 --> 00:01:50,000 I'll also discuss about what are this foundation models. 36 00:01:50,000 --> 00:01:55,000 As I said, so many different different companies are coming up with some LLM models instead. 37 00:01:55,000 --> 00:01:57,000 So that is the foundation model. 38 00:01:57,000 --> 00:01:58,000 You can do the fine tuning here itself. 39 00:01:58,000 --> 00:02:04,000 You can use any of these models, perform any task, and even directly you can use it in your application 40 00:02:04,000 --> 00:02:05,000 wherever it is required. 41 00:02:05,000 --> 00:02:12,000 Okay, so if I talk about, uh, Amazon Bedrock, it is fully managed service that makes FMS. 42 00:02:12,000 --> 00:02:16,000 That is foundation models from leading AI startup and Amazon available via an API. 43 00:02:16,000 --> 00:02:22,000 So you can choose from a wide range of FMS to find the model that is best suited for your case. 44 00:02:22,000 --> 00:02:25,000 So with bedrock serverless experience, you can start quickly. 45 00:02:25,000 --> 00:02:31,000 Privately customize a FME Foundation model with your own data and easily integrate and deploy them in 46 00:02:31,000 --> 00:02:33,000 your application using AWS tools. 47 00:02:33,000 --> 00:02:38,000 So here you don't have to even worry about the deployment, the scalability, and everything as such. 48 00:02:38,000 --> 00:02:42,000 So, uh, let me just go ahead and click on Get Started. 49 00:02:42,000 --> 00:02:46,000 So this is how it looks like now what all companies it specifically support. 50 00:02:46,000 --> 00:02:48,000 Because I'm also going to show you the entire coding also. 51 00:02:48,000 --> 00:02:52,000 Right how you can use all these models right now in this foundation model. 52 00:02:52,000 --> 00:02:55,000 Uh, there are different different models like from AI 21 labs. 53 00:02:55,000 --> 00:02:58,000 You can see Jurassic two series is there from Amazon. 54 00:02:58,000 --> 00:02:59,000 You have Titan models. 55 00:02:59,000 --> 00:03:03,000 From cloudy you have this cloudy models by Andrew Anthropic. 56 00:03:03,000 --> 00:03:04,000 Uh, you have llama two. 57 00:03:04,000 --> 00:03:08,000 You have stable diffusion by stability AI you have this command by cohere. 58 00:03:08,000 --> 00:03:13,000 So this through this you will be able to perform different different use cases like chat text image. 59 00:03:13,000 --> 00:03:14,000 Right. 60 00:03:14,000 --> 00:03:16,000 And it also provides you hands on lab. 61 00:03:16,000 --> 00:03:20,000 Uh along with that some kind of learning curves, some basic learning course, which you can probably 62 00:03:20,000 --> 00:03:21,000 get started to. 63 00:03:21,000 --> 00:03:23,000 Now here are some of the examples. 64 00:03:23,000 --> 00:03:28,000 Here you can see what all, uh, different different use cases you can solve with respect to different 65 00:03:28,000 --> 00:03:28,000 different models. 66 00:03:28,000 --> 00:03:29,000 Right. 67 00:03:29,000 --> 00:03:33,000 So with the help of Titan text G1, you can see action items from a meeting. 68 00:03:33,000 --> 00:03:33,000 Transcript. 69 00:03:33,000 --> 00:03:35,000 Advanced Q&A with citation. 70 00:03:36,000 --> 00:03:38,000 Uh with llama two chat 13 B right. 71 00:03:38,000 --> 00:03:41,000 This 13 billion parameters you have you can create chain of thoughts. 72 00:03:41,000 --> 00:03:46,000 Then here also you have with 70 billion parameters with cloud you can have character load, role play, 73 00:03:47,000 --> 00:03:51,000 code generation, content generation, contract entity extraction. 74 00:03:51,000 --> 00:03:53,000 Create an image, create an image. 75 00:03:53,000 --> 00:03:55,000 Here you can see creating a table of product description. 76 00:03:56,000 --> 00:04:00,000 Uh, then you have debug code with llama 230 uh 13 billion parameters llama two chat. 77 00:04:00,000 --> 00:04:03,000 Then you have Jurassic, then you have this right. 78 00:04:03,000 --> 00:04:07,000 So as soon as you probably click on anything, let's say I want to go and see that how it is going to 79 00:04:07,000 --> 00:04:08,000 create an image. 80 00:04:08,000 --> 00:04:10,000 So here it is, what it looks like, right? 81 00:04:10,000 --> 00:04:13,000 This is the entire API request that you have to probably call. 82 00:04:13,000 --> 00:04:19,000 And with the help of this this image will be generated let's say I will probably let's see here okay 83 00:04:19,000 --> 00:04:23,000 I can open this in playground and I can give my own. 84 00:04:23,000 --> 00:04:29,000 So HD image of a beach. 85 00:04:29,000 --> 00:04:34,000 So this is the playground where you can with a sunset. 86 00:04:34,000 --> 00:04:38,000 With sunrise, I can basically write like this. 87 00:04:38,000 --> 00:04:38,000 Right. 88 00:04:38,000 --> 00:04:42,000 So as soon as I give this prompt and if I run it, you will be able to see that I'll be able to see 89 00:04:42,000 --> 00:04:43,000 this entire image. 90 00:04:43,000 --> 00:04:45,000 So it actually creates this. 91 00:04:45,000 --> 00:04:47,000 This is from the stable diffusion itself. 92 00:04:47,000 --> 00:04:47,000 Right. 93 00:04:47,000 --> 00:04:51,000 So here you'll be able to see once we wait, it, uh, we'll be able to run it. 94 00:04:51,000 --> 00:04:54,000 And now here you can see this is an amazing image that is created. 95 00:04:54,000 --> 00:04:59,000 And you can use HD image, cinematic display, all different, different things you can probably put 96 00:04:59,000 --> 00:05:00,000 over here. 97 00:05:00,000 --> 00:05:03,000 Uh, I still go with respect to the overviews. 98 00:05:03,000 --> 00:05:06,000 So as I showed all this is there are there examples. 99 00:05:06,000 --> 00:05:08,000 You can probably see all these things are there. 100 00:05:08,000 --> 00:05:10,000 You can go ahead and check out in the playground. 101 00:05:10,000 --> 00:05:13,000 Let's say there is something called as creating table or product description. 102 00:05:13,000 --> 00:05:16,000 Here you can see this is the prompt right. 103 00:05:16,000 --> 00:05:18,000 Sunglasses keywords polarized is this is there. 104 00:05:18,000 --> 00:05:20,000 Let's say this is your table. 105 00:05:20,000 --> 00:05:25,000 You can probably ask any kind of questions and get with respect to the answers also over here right. 106 00:05:25,000 --> 00:05:26,000 So different different use cases is there. 107 00:05:26,000 --> 00:05:28,000 At the end of the day I'll just not show you this APIs. 108 00:05:28,000 --> 00:05:31,000 Now let's go ahead and implement all these things right. 109 00:05:31,000 --> 00:05:36,000 And there steps, some steps which uh we will be seeing completely from scratch again. 110 00:05:36,000 --> 00:05:39,000 Uh, you can use any of these services as such. 111 00:05:39,000 --> 00:05:39,000 Yes. 112 00:05:39,000 --> 00:05:43,000 There is some pricing because at the end of the day you're using some cloud services, right? 113 00:05:43,000 --> 00:05:46,000 It always depends on the tokens that you are specifically getting. 114 00:05:46,000 --> 00:05:49,000 So based on that, Amazon bedrock basically charges you. 115 00:05:49,000 --> 00:05:54,000 So if you want to see the pricing you can probably go ahead over here and right Amazon Bedrock. 116 00:05:54,000 --> 00:05:56,000 And you can click on pricing right. 117 00:05:56,000 --> 00:06:00,000 So bedrock pricing here is the link. 118 00:06:00,000 --> 00:06:02,000 Here you'll be able to see the pricing overview. 119 00:06:02,000 --> 00:06:04,000 Different different pricing is there. 120 00:06:04,000 --> 00:06:07,000 And this is 4000 input tokens right. 121 00:06:07,000 --> 00:06:08,000 Price per thousand. 122 00:06:08,000 --> 00:06:10,000 Output tokens, input tokens output tokens. 123 00:06:10,000 --> 00:06:12,000 So all these prices is there. 124 00:06:12,000 --> 00:06:14,000 You can compare it with OpenAI models and all. 125 00:06:14,000 --> 00:06:18,000 But still this pricing will get reduced as we go ahead. 126 00:06:18,000 --> 00:06:19,000 More custom solutions will probably come up. 127 00:06:19,000 --> 00:06:20,000 Right. 128 00:06:20,000 --> 00:06:22,000 So this is they're all good things. 129 00:06:22,000 --> 00:06:26,000 And you'll also be able to understand uh, see Titan is also having this multi model embeddings. 130 00:06:26,000 --> 00:06:31,000 But at the end of the day we should know that how we can actually create entirely, completely from 131 00:06:31,000 --> 00:06:32,000 scratch. 132 00:06:32,000 --> 00:06:38,000 So let me go ahead and start, uh, this let's see whether I've opened any VSCode or not. 133 00:06:38,000 --> 00:06:44,000 So I will go ahead and start my VSCode. 134 00:06:46,000 --> 00:06:49,000 Um, so let's see over here. 135 00:06:49,000 --> 00:06:51,000 Here is our folder that I've created over here. 136 00:06:51,000 --> 00:06:56,000 Let me rename this to, uh, AWS bedrock. 137 00:06:56,000 --> 00:06:57,000 Okay. 138 00:06:57,000 --> 00:06:59,000 And let me show you. 139 00:06:59,000 --> 00:07:00,000 Okay. 140 00:07:01,000 --> 00:07:03,000 Um, bedrock. 141 00:07:03,000 --> 00:07:03,000 Okay. 142 00:07:03,000 --> 00:07:07,000 And, uh, let me just quickly show you how you can probably start it. 143 00:07:07,000 --> 00:07:07,000 Okay? 144 00:07:07,000 --> 00:07:13,000 So I'll show you the entire setup from starting and all what all things is basically required over here. 145 00:07:13,000 --> 00:07:16,000 And, uh, you also require an IAM key over here itself. 146 00:07:16,000 --> 00:07:17,000 Right. 147 00:07:17,000 --> 00:07:19,000 So first of all, I will open the terminal. 148 00:07:19,000 --> 00:07:23,000 Now after opening the terminal over here, I will open the command prompt. 149 00:07:23,000 --> 00:07:24,000 As usual. 150 00:07:24,000 --> 00:07:27,000 We will go ahead and create a new environment. 151 00:07:27,000 --> 00:07:28,000 That is always a good idea. 152 00:07:28,000 --> 00:07:29,000 Okay. 153 00:07:29,000 --> 00:07:39,000 So for creating a new environment I'll write conda create minus p v n v python double equal to 3.10 154 00:07:39,000 --> 00:07:40,000 with y. 155 00:07:40,000 --> 00:07:41,000 Okay. 156 00:07:41,000 --> 00:07:42,000 So I hope everybody knows about this. 157 00:07:42,000 --> 00:07:45,000 So this will basically create my V and v environment. 158 00:07:45,000 --> 00:07:48,000 And once that environment is created I will activate it. 159 00:07:48,000 --> 00:07:53,000 Till then I will also go ahead and install requirements dot txt. 160 00:07:54,000 --> 00:07:59,000 Okay so once I install requirement dot txt I will be requiring some libraries like Boto3. 161 00:07:59,000 --> 00:08:04,000 So Boto3 is a library which will actually help me to connect all the services that we have in Amazon 162 00:08:04,000 --> 00:08:05,000 bedrock. 163 00:08:05,000 --> 00:08:05,000 Right. 164 00:08:05,000 --> 00:08:07,000 Uh, AWS bedrock also. 165 00:08:07,000 --> 00:08:13,000 So this is some of the things, uh, along with this, uh, I will also go ahead and import or install 166 00:08:14,000 --> 00:08:14,000 AWS CLI. 167 00:08:14,000 --> 00:08:15,000 Right. 168 00:08:15,000 --> 00:08:19,000 So guys, now after creating the environment I will go ahead and activate the environment. 169 00:08:19,000 --> 00:08:21,000 So V and V. 170 00:08:21,000 --> 00:08:23,000 So my environment has been activated. 171 00:08:23,000 --> 00:08:28,000 Now what I'm actually going to do over here I'll save this requirement dot txt. 172 00:08:28,000 --> 00:08:33,000 And then we will go ahead and install these two libraries. 173 00:08:33,000 --> 00:08:39,000 One is pip install pip install minus our requirement dot txt. 174 00:08:39,000 --> 00:08:44,000 So once I specifically install all these requirements one is two two only libraries are required Boto3 175 00:08:44,000 --> 00:08:46,000 and aws cli. 176 00:08:46,000 --> 00:08:54,000 Now the next thing till this requirement is getting installed, we also need to create and IAM user 177 00:08:54,000 --> 00:08:58,000 so that we will be able to configure it with the AWS itself. 178 00:08:58,000 --> 00:08:58,000 Right. 179 00:08:58,000 --> 00:09:06,000 So what I'm actually going to go I'll go and click on the home page okay I will sign into the console 180 00:09:07,000 --> 00:09:07,000 again. 181 00:09:08,000 --> 00:09:13,000 And here I will search for IAM user okay. 182 00:09:13,000 --> 00:09:20,000 So once you go to the IAM user you will be able to see that I will go ahead and create a user itself. 183 00:09:20,000 --> 00:09:22,000 So right now there are multiple users. 184 00:09:22,000 --> 00:09:24,000 So let me do one thing. 185 00:09:24,000 --> 00:09:25,000 Let me go and see the user already. 186 00:09:25,000 --> 00:09:33,000 Krish is created but I will just create one more user let's say test test admin okay so this is the 187 00:09:33,000 --> 00:09:36,000 user that I will specifically create and I will click on. 188 00:09:36,000 --> 00:09:39,000 Next I will say attach policies directly. 189 00:09:39,000 --> 00:09:43,000 I will select this option and I will give the administrator access okay. 190 00:09:43,000 --> 00:09:47,000 So administrator access basically means that we are giving the administrator access itself. 191 00:09:47,000 --> 00:09:48,000 Right. 192 00:09:48,000 --> 00:09:52,000 But again when you are working in the company you will definitely not get this access. 193 00:09:52,000 --> 00:09:56,000 Uh, instead, uh, based on the services that you are specifically using, that access you will get. 194 00:09:56,000 --> 00:09:57,000 Okay. 195 00:09:57,000 --> 00:10:00,000 Um, I will go ahead and create the policy. 196 00:10:01,000 --> 00:10:06,000 So let's see, once I go over here, then it will ask for some permission. 197 00:10:08,000 --> 00:10:11,000 So okay, forget about policies. 198 00:10:11,000 --> 00:10:12,000 Then I will go ahead. 199 00:10:12,000 --> 00:10:16,000 And here you will be able to see that I'll select the next button okay. 200 00:10:17,000 --> 00:10:21,000 After selecting the next button you can see I'm getting the administrator access along with this. 201 00:10:21,000 --> 00:10:22,000 What I will go I'll do. 202 00:10:22,000 --> 00:10:25,000 Go ahead and create the user itself. 203 00:10:25,000 --> 00:10:29,000 Now this is the user that you can see over here test admin right. 204 00:10:29,000 --> 00:10:32,000 And now let me click on this Create access key. 205 00:10:32,000 --> 00:10:32,000 Right. 206 00:10:32,000 --> 00:10:35,000 So I definitely require an access key. 207 00:10:35,000 --> 00:10:37,000 So for that I will go ahead and create this. 208 00:10:38,000 --> 00:10:40,000 Over here you can select Command Line interface. 209 00:10:40,000 --> 00:10:45,000 Since we are going to use it for CLI I will go ahead and click on this I understand and then I will 210 00:10:45,000 --> 00:10:47,000 go ahead and click on the next. 211 00:10:47,000 --> 00:10:51,000 Once I click on the next, you will be able to see that some description value it will ask for. 212 00:10:51,000 --> 00:10:53,000 I will say test key okay. 213 00:10:53,000 --> 00:10:55,000 And let me go ahead and create the access key. 214 00:10:55,000 --> 00:10:59,000 So here you can see I have the access key I have the secret access key. 215 00:10:59,000 --> 00:11:03,000 Now what you can do is that in this particular case you can download this. 216 00:11:03,000 --> 00:11:03,000 Right. 217 00:11:03,000 --> 00:11:06,000 So that later on you will not be able to see this. 218 00:11:06,000 --> 00:11:09,000 So it is a good idea that you download this in the form of CSV file. 219 00:11:09,000 --> 00:11:12,000 So I'll download this in the form of CSV file. 220 00:11:12,000 --> 00:11:14,000 And inside this CSV file I'll get this two values. 221 00:11:14,000 --> 00:11:17,000 One is access key, one is the secret access key. 222 00:11:17,000 --> 00:11:20,000 So let me quickly go ahead and copy this. 223 00:11:20,000 --> 00:11:22,000 And now I will go ahead and configure it. 224 00:11:22,000 --> 00:11:22,000 Right. 225 00:11:22,000 --> 00:11:24,000 We need to configure it over here. 226 00:11:25,000 --> 00:11:30,000 So already you know that we have also made sure that this AWS CLI is available over here. 227 00:11:30,000 --> 00:11:31,000 And it is installed. 228 00:11:31,000 --> 00:11:36,000 Now what we are going to do quickly is that we will go ahead and configure it from the command line. 229 00:11:36,000 --> 00:11:39,000 So let me go ahead and write AWS configure. 230 00:11:39,000 --> 00:11:39,000 Right. 231 00:11:39,000 --> 00:11:43,000 So here you will be able to see it is asking for AWS key ID. 232 00:11:43,000 --> 00:11:45,000 So I will copy and paste it over here. 233 00:11:45,000 --> 00:11:48,000 So I will copy this and I will paste it over here. 234 00:11:48,000 --> 00:11:49,000 Okay. 235 00:11:49,000 --> 00:11:54,000 So once I probably paste it you will be able to see that it will also ask for secret access key. 236 00:11:54,000 --> 00:11:56,000 So here I will copy it. 237 00:11:56,000 --> 00:12:00,000 And let me quickly go ahead and paste it over here. 238 00:12:00,000 --> 00:12:02,000 So here is what I see right now. 239 00:12:02,000 --> 00:12:03,000 You are able to see the access key. 240 00:12:03,000 --> 00:12:07,000 But don't worry, I will delete this as soon as I complete this video. 241 00:12:07,000 --> 00:12:10,000 Now the default region it is basically asking. 242 00:12:10,000 --> 00:12:11,000 So right now it is US East one. 243 00:12:11,000 --> 00:12:16,000 So if I go ahead and see my Amazon bedrock right now where I am pointing to. 244 00:12:16,000 --> 00:12:18,000 So please make sure that you point it to us East one. 245 00:12:18,000 --> 00:12:22,000 If you are not pointing over here, you can basically whatever region you are pointing, you have to 246 00:12:22,000 --> 00:12:23,000 write it over there. 247 00:12:23,000 --> 00:12:26,000 But right now, according to me, I'm going to point it over here, right? 248 00:12:26,000 --> 00:12:29,000 That is with respect to us East one. 249 00:12:29,000 --> 00:12:34,000 Now, one more additional setting you really need to do is that go to the model access. 250 00:12:34,000 --> 00:12:39,000 Because see, initially when you are in this US East one, this model access will not be given. 251 00:12:40,000 --> 00:12:40,000 Right. 252 00:12:40,000 --> 00:12:45,000 So over here you have to access you need to have the status of access granted. 253 00:12:45,000 --> 00:12:47,000 Then only you'll be able to use this specific models. 254 00:12:47,000 --> 00:12:53,000 Now for that when you click on Manage Model access right automatically the selection will come. 255 00:12:53,000 --> 00:12:56,000 And here you will be able to just save the changes here. 256 00:12:56,000 --> 00:12:58,000 One option off activate changes will come. 257 00:12:58,000 --> 00:13:01,000 See in the bottom side like save changes is over here. 258 00:13:01,000 --> 00:13:03,000 So here I've enabled it. 259 00:13:03,000 --> 00:13:05,000 And this access is granted right now. 260 00:13:05,000 --> 00:13:09,000 If it is not granted then this all will be disabled right now. 261 00:13:09,000 --> 00:13:10,000 Right. 262 00:13:10,000 --> 00:13:15,000 So what I will do, just to show you an example, let me just reload this page. 263 00:13:15,000 --> 00:13:17,000 Let me show you whether I have granted some other. 264 00:13:18,000 --> 00:13:25,000 See let's say with respect to Singapore if I'm in Asia Pacific, Singapore region right now, inside 265 00:13:25,000 --> 00:13:27,000 the Singapore region, you will be able to see that. 266 00:13:28,000 --> 00:13:31,000 See, something is not available over here. 267 00:13:31,000 --> 00:13:31,000 Right. 268 00:13:31,000 --> 00:13:31,000 Like this. 269 00:13:31,000 --> 00:13:32,000 Two are not available. 270 00:13:32,000 --> 00:13:34,000 So I will skip Singapore. 271 00:13:34,000 --> 00:13:36,000 Let's go to Tokyo. 272 00:13:36,000 --> 00:13:36,000 Okay. 273 00:13:37,000 --> 00:13:41,000 Just to show you an idea, because I have already activated in US East one. 274 00:13:41,000 --> 00:13:47,000 So if I probably see with this region here, you you'll be able to see this much excess you have available 275 00:13:47,000 --> 00:13:49,000 to a request available to request. 276 00:13:49,000 --> 00:13:55,000 So if I probably click on manage model and if I probably just click on this, let's say these are available 277 00:13:55,000 --> 00:13:55,000 right. 278 00:13:55,000 --> 00:13:58,000 I just need to click on Request Model access. 279 00:13:58,000 --> 00:14:01,000 So for this region only this three models are available. 280 00:14:01,000 --> 00:14:03,000 This two models are available. 281 00:14:03,000 --> 00:14:09,000 So my always suggestion would be that you go to US East one right in US US East one You will be able 282 00:14:09,000 --> 00:14:11,000 to see multiple models like this, right? 283 00:14:11,000 --> 00:14:13,000 Multiple models, you'll be able to see it. 284 00:14:14,000 --> 00:14:14,000 Okay. 285 00:14:14,000 --> 00:14:19,000 So you have to request for the access right. 286 00:14:19,000 --> 00:14:22,000 If I probably you can select us East one also. 287 00:14:22,000 --> 00:14:24,000 Or you can also select US West. 288 00:14:24,000 --> 00:14:25,000 It is up to you. 289 00:14:25,000 --> 00:14:26,000 US West two. 290 00:14:26,000 --> 00:14:29,000 So here also they provide you the access of all the models. 291 00:14:29,000 --> 00:14:29,000 Right. 292 00:14:29,000 --> 00:14:32,000 So understand one thing why I'm showing you all these things. 293 00:14:32,000 --> 00:14:35,000 Because initially you really need to request access. 294 00:14:35,000 --> 00:14:37,000 And this is also what I have requested already. 295 00:14:37,000 --> 00:14:40,000 So here you'll be able to find out more use cases. 296 00:14:40,000 --> 00:14:40,000 Right. 297 00:14:40,000 --> 00:14:42,000 So everything is available right. 298 00:14:42,000 --> 00:14:48,000 But by default I will go ahead and use this US east of N Virginia okay. 299 00:14:48,000 --> 00:14:51,000 So right now again my access is granted. 300 00:14:51,000 --> 00:14:55,000 So you if your access is not granted you have to request for it okay. 301 00:14:55,000 --> 00:15:00,000 So once this is done I will go back to my command prompt here. 302 00:15:00,000 --> 00:15:02,000 The default region is US East one. 303 00:15:02,000 --> 00:15:03,000 So I'll press enter. 304 00:15:03,000 --> 00:15:04,000 This is what I really want. 305 00:15:04,000 --> 00:15:07,000 And default output format I want in the form of JSON. 306 00:15:07,000 --> 00:15:09,000 So I will go ahead and write JSON. 307 00:15:09,000 --> 00:15:09,000 Okay. 308 00:15:09,000 --> 00:15:11,000 So this is the initial setup you really need to do. 309 00:15:12,000 --> 00:15:15,000 Now let's start working on different different models. 310 00:15:15,000 --> 00:15:23,000 So so guys now once we have actually configured now we will go ahead and use any of these models as 311 00:15:23,000 --> 00:15:25,000 an example and we'll try to perform various tasks. 312 00:15:25,000 --> 00:15:28,000 So in this example uh, let me do one thing. 313 00:15:28,000 --> 00:15:33,000 Uh, let me take an example of content generation in cloud. 314 00:15:33,000 --> 00:15:33,000 Okay. 315 00:15:34,000 --> 00:15:41,000 So cloudy cloudy one we can specifically take or uh, let's say I want llama two, you can actually 316 00:15:41,000 --> 00:15:41,000 take llama two. 317 00:15:41,000 --> 00:15:43,000 So it's it's up to you. 318 00:15:43,000 --> 00:15:46,000 So let's first of all start start with this llama two itself. 319 00:15:46,000 --> 00:15:52,000 And for this I'm going to specifically use my Lama 270 billion parameters okay. 320 00:15:52,000 --> 00:15:53,000 The chain of thoughts okay. 321 00:15:53,000 --> 00:15:59,000 And with respect to every model you will be getting some model ID content type. 322 00:15:59,000 --> 00:16:01,000 So this is how is the API request needs to go. 323 00:16:01,000 --> 00:16:04,000 So we'll form this entire API request through our code. 324 00:16:04,000 --> 00:16:06,000 And then we will go ahead and hit this okay. 325 00:16:06,000 --> 00:16:10,000 So let me quickly go ahead and start my coding okay. 326 00:16:10,000 --> 00:16:13,000 So here I will rename this to llama two. 327 00:16:13,000 --> 00:16:16,000 The llama two. 328 00:16:16,000 --> 00:16:19,000 So here is the file guys llama 2.py. 329 00:16:19,000 --> 00:16:21,000 So I will start my code over here. 330 00:16:21,000 --> 00:16:25,000 So first of all let me go ahead and write import Boto3 I'm going to import Boto3 along with it. 331 00:16:25,000 --> 00:16:27,000 I'm also going to import JSON. 332 00:16:28,000 --> 00:16:35,000 Uh the first thing as usual you will be able to see that whenever I see this file writes a chain of 333 00:16:35,000 --> 00:16:40,000 thought over here, I have to create my API request with the help of model name model ID. 334 00:16:40,000 --> 00:16:44,000 Then I need to also put content type, then apply accept. 335 00:16:44,000 --> 00:16:49,000 Then my body should look something like this prompt with all this information. 336 00:16:49,000 --> 00:16:50,000 See uh in lambda two. 337 00:16:50,000 --> 00:16:56,000 So basically your uh your uh prompt uh intro will start with or prompt. 338 00:16:56,000 --> 00:17:00,000 First character will probably start with first word will start with this okay. 339 00:17:00,000 --> 00:17:05,000 It's just like an instruction okay I insist then here you have multiple options. 340 00:17:05,000 --> 00:17:07,000 You can probably see this right here. 341 00:17:07,000 --> 00:17:13,000 Uh you have options like max gen length then temperature, then top P 0.9, something like this okay. 342 00:17:13,000 --> 00:17:17,000 So top P basically means I think it is with respect to the probability. 343 00:17:17,000 --> 00:17:21,000 So let's quickly go ahead and set up that entire command prompt. 344 00:17:21,000 --> 00:17:24,000 So what I will do is that I will copy this entirely okay. 345 00:17:24,000 --> 00:17:28,000 And I will create one JSON okay. 346 00:17:28,000 --> 00:17:33,000 I will write test dot JSON so that I have that format over here. 347 00:17:33,000 --> 00:17:35,000 So this is the JSON. 348 00:17:35,000 --> 00:17:39,000 I have to make sure that I have to put all my API requests in this form. 349 00:17:39,000 --> 00:17:39,000 Okay. 350 00:17:39,000 --> 00:17:40,000 For using Lambda two. 351 00:17:40,000 --> 00:17:45,000 Now quickly I will go ahead and write one prompt underscore data. 352 00:17:45,000 --> 00:17:49,000 I will say, hey, uh, this prompt data, I want to do something like this. 353 00:17:49,000 --> 00:17:50,000 I'll act. 354 00:17:50,000 --> 00:18:01,000 I'll say act as this as, uh, Shakespeare and write a poem on machine learning. 355 00:18:01,000 --> 00:18:05,000 Let's say I this is my prompt that I specifically want to use. 356 00:18:05,000 --> 00:18:05,000 Okay. 357 00:18:05,000 --> 00:18:10,000 Now, the most amazing thing is that Amazon AWS bedrock is quite good. 358 00:18:10,000 --> 00:18:12,000 You know, it provides you easy APIs. 359 00:18:12,000 --> 00:18:16,000 The APIs request, you just need to use it and, uh, start serving it. 360 00:18:16,000 --> 00:18:17,000 Start using it in your application. 361 00:18:17,000 --> 00:18:21,000 So I will first of all go ahead and use this bedrock. 362 00:18:21,000 --> 00:18:25,000 So I will write bedrock is equal to Boto3 dot client. 363 00:18:26,000 --> 00:18:30,000 And then I will specifically use my service underscore name. 364 00:18:30,000 --> 00:18:33,000 Uh service underscore name is equal to. 365 00:18:33,000 --> 00:18:37,000 And I have to give the service name as bedrock or dash runtime. 366 00:18:37,000 --> 00:18:39,000 So this is the first thing that we really need to do. 367 00:18:39,000 --> 00:18:41,000 This is the client name. 368 00:18:41,000 --> 00:18:44,000 Like this is the service name that we really need to use. 369 00:18:44,000 --> 00:18:47,000 And with the help of Boto3, we'll be able to connect to the bedrock itself. 370 00:18:47,000 --> 00:18:49,000 So this is my bedrock over here. 371 00:18:49,000 --> 00:18:53,000 Now the next thing is that I need to form the payload. 372 00:18:53,000 --> 00:18:53,000 Right. 373 00:18:53,000 --> 00:18:56,000 So this will basically be my payload structure as a dictionary. 374 00:18:56,000 --> 00:18:59,000 I will try to find give the values in key value pair. 375 00:18:59,000 --> 00:19:02,000 As usual, if I probably see the test duration. 376 00:19:02,000 --> 00:19:03,000 This is what is my body. 377 00:19:03,000 --> 00:19:06,000 So I need to have in the form of key value pairs. 378 00:19:06,000 --> 00:19:08,000 Prompt is equal to some value. 379 00:19:08,000 --> 00:19:15,000 Then you'll be able to see max gen length, some value temperature, some value top underscore p some 380 00:19:15,000 --> 00:19:15,000 value. 381 00:19:15,000 --> 00:19:18,000 So similarly I will also give my payload in this format. 382 00:19:18,000 --> 00:19:21,000 So here I will write prompt colon. 383 00:19:21,000 --> 00:19:26,000 And here I will go ahead and write I honestly since this is my intro right. 384 00:19:26,000 --> 00:19:33,000 And uh, this intro will be concatenated with my prompt underscore data. 385 00:19:33,000 --> 00:19:34,000 Right. 386 00:19:34,000 --> 00:19:37,000 So this I'm concatenating with my prompt underscore data. 387 00:19:37,000 --> 00:19:42,000 And in the end I have to probably also end this instruction okay. 388 00:19:42,000 --> 00:19:46,000 So this is how my prompt this this prompt body has got created right. 389 00:19:46,000 --> 00:19:48,000 This this is what it has got created. 390 00:19:48,000 --> 00:19:48,000 Right. 391 00:19:48,000 --> 00:19:50,000 Because the prompt will get inserted over here. 392 00:19:50,000 --> 00:19:55,000 Now along with this prompt I need to provide some more different different values. 393 00:19:55,000 --> 00:19:55,000 Right. 394 00:19:55,000 --> 00:19:57,000 What all values I need to provide. 395 00:19:57,000 --> 00:20:02,000 One example is that I'm getting from my API max gen underscore length okay. 396 00:20:02,000 --> 00:20:04,000 And this will also be in the form of key value pairs. 397 00:20:04,000 --> 00:20:07,000 And here I will mention phi to L okay. 398 00:20:07,000 --> 00:20:16,000 And then the next thing is that I will go ahead and write temperature with colon 0.5 comma. 399 00:20:18,000 --> 00:20:23,000 Top underscore p colon 0.9 okay. 400 00:20:23,000 --> 00:20:26,000 So these are my initial values that I've actually set up. 401 00:20:26,000 --> 00:20:30,000 And this is not thing but this is my payload the payload that is specifically going over here. 402 00:20:30,000 --> 00:20:35,000 So once we do this the next thing what I will do this is basically my body. 403 00:20:35,000 --> 00:20:42,000 So I will write body json.dumps I will basically convert this into a JSON and I will use my payload 404 00:20:42,000 --> 00:20:43,000 over here. 405 00:20:43,000 --> 00:20:43,000 Okay. 406 00:20:43,000 --> 00:20:48,000 So once my body is basically created the next thing what I'm actually going to do over here, I'm going 407 00:20:48,000 --> 00:20:53,000 to basically write my model ID and your model ID name. 408 00:20:53,000 --> 00:20:56,000 As usual, the model ID name is this one meta. 409 00:20:56,000 --> 00:20:58,000 This one version one something. 410 00:20:58,000 --> 00:20:58,000 Okay. 411 00:20:58,000 --> 00:20:59,000 So I'll paste it over here. 412 00:20:59,000 --> 00:21:02,000 So this basically becomes my model ID itself. 413 00:21:02,000 --> 00:21:04,000 Now I'll go ahead and create my response. 414 00:21:04,000 --> 00:21:09,000 Now in that response what I need to do I just need to write bedrock, the bedrock object that is created 415 00:21:09,000 --> 00:21:15,000 over here dot invoke invoke underscore model okay. 416 00:21:15,000 --> 00:21:18,000 Now inside this I will give my first parameter. 417 00:21:18,000 --> 00:21:20,000 The first parameter is nothing but body. 418 00:21:20,000 --> 00:21:23,000 The body will be initialized to this specific body itself. 419 00:21:23,000 --> 00:21:30,000 Uh, whatever body we have actually created, the second parameter is nothing but my model ID, so model. 420 00:21:30,000 --> 00:21:32,000 Oops, just a second bedrock. 421 00:21:32,000 --> 00:21:38,000 So I have to use this model id okay. 422 00:21:38,000 --> 00:21:42,000 So this will basically be my second parameter that I really need to give model ID. 423 00:21:42,000 --> 00:21:49,000 And this I will initialize to what we need to initialize to this model ID, right? 424 00:21:49,000 --> 00:21:51,000 So my second parameter is also done. 425 00:21:51,000 --> 00:21:51,000 Body is done. 426 00:21:51,000 --> 00:21:53,000 Model ID is done. 427 00:21:53,000 --> 00:21:57,000 After model id I need to give my accept token as usual. 428 00:21:57,000 --> 00:22:03,000 What is the accept token in test.js on it is nothing but application slash JSON, so I'll paste it over 429 00:22:03,000 --> 00:22:03,000 here. 430 00:22:04,000 --> 00:22:04,000 Oops. 431 00:22:05,000 --> 00:22:06,000 I'll paste it over here. 432 00:22:06,000 --> 00:22:07,000 Done. 433 00:22:07,000 --> 00:22:12,000 The next parameter after this will be nothing, but let me see. 434 00:22:12,000 --> 00:22:16,000 Content type okay, so I have to also give my content type over here. 435 00:22:16,000 --> 00:22:19,000 This will also be my application slash JSON I guess. 436 00:22:20,000 --> 00:22:23,000 Again I'm just seeing this API, so whatever values are there my body is ready. 437 00:22:23,000 --> 00:22:26,000 My accept model ID content type, everything is ready. 438 00:22:26,000 --> 00:22:29,000 These are the four parameters I need to give in my invoke model, right? 439 00:22:29,000 --> 00:22:31,000 What model I'm specifically invoking. 440 00:22:31,000 --> 00:22:35,000 Now, once I get this, uh, you will be able to see that once. 441 00:22:35,000 --> 00:22:39,000 I probably go ahead and write response underscore. 442 00:22:39,000 --> 00:22:42,000 This will basically be my response inside this response. 443 00:22:42,000 --> 00:22:43,000 Understand? 444 00:22:43,000 --> 00:22:46,000 Once I get this specific response, I'll write. 445 00:22:46,000 --> 00:22:48,000 I'll take the body part okay. 446 00:22:48,000 --> 00:22:52,000 So inner part, uh, because there'll be a lot many different different contents that will be coming 447 00:22:52,000 --> 00:22:52,000 up. 448 00:22:52,000 --> 00:22:58,000 So if I write JSON dot loads I will take this value and I will say response dot get. 449 00:22:58,000 --> 00:23:02,000 So inside this there will be a key which will be called as body okay. 450 00:23:02,000 --> 00:23:07,000 Inside that body you will find out entire information of the response that you are able to get. 451 00:23:07,000 --> 00:23:12,000 In this particular case, whatever text I'm writing, act as a secretary and write a poem on machine 452 00:23:12,000 --> 00:23:12,000 learning. 453 00:23:12,000 --> 00:23:16,000 So this is basically my prompt and it will give that specific text inside this body. 454 00:23:16,000 --> 00:23:16,000 Okay. 455 00:23:16,000 --> 00:23:18,000 So I will write this. 456 00:23:18,000 --> 00:23:18,000 Okay. 457 00:23:19,000 --> 00:23:21,000 Um, this body dot read this is done. 458 00:23:21,000 --> 00:23:23,000 Now I will go ahead and print. 459 00:23:23,000 --> 00:23:25,000 I can also print it, but I don't want it. 460 00:23:25,000 --> 00:23:28,000 So let me go ahead and write response underscore text. 461 00:23:28,000 --> 00:23:33,000 Now inside this one thing that you will be able to see that whenever we use llama two. 462 00:23:33,000 --> 00:23:38,000 If you don't know, I've also created videos with respect to llama two, the open source model. 463 00:23:38,000 --> 00:23:41,000 If I use this response underscore body, they will also be a key. 464 00:23:41,000 --> 00:23:44,000 See, inside this body there'll be a body key. 465 00:23:44,000 --> 00:23:49,000 Inside that there will be another key which will be called as generation which will have the entire 466 00:23:49,000 --> 00:23:51,000 text of the response that we want. 467 00:23:51,000 --> 00:23:52,000 Okay. 468 00:23:52,000 --> 00:23:56,000 So I will just go ahead and print this response. 469 00:23:56,000 --> 00:23:57,000 Underscore text. 470 00:23:59,000 --> 00:24:00,000 Done. 471 00:24:01,000 --> 00:24:07,000 Now let's see if everything is working fine or not and whether we will be able to see each and everything. 472 00:24:07,000 --> 00:24:08,000 So this is very simple. 473 00:24:08,000 --> 00:24:09,000 A prompt is over here. 474 00:24:09,000 --> 00:24:11,000 We have invoked the bedrock. 475 00:24:11,000 --> 00:24:13,000 Then there is a payload. 476 00:24:13,000 --> 00:24:18,000 Then here you have this entire body, uh, body I've created in the form of JSON. 477 00:24:18,000 --> 00:24:20,000 Then model ID accept content type. 478 00:24:20,000 --> 00:24:23,000 The main thing is understand this test JSON right. 479 00:24:23,000 --> 00:24:24,000 And then you will be able to do it. 480 00:24:24,000 --> 00:24:25,000 So perfect. 481 00:24:25,000 --> 00:24:31,000 Let's execute this now and let's see whether everything is working fine or not okay. 482 00:24:31,000 --> 00:24:37,000 So I will just go ahead and write Python Llama two. 483 00:24:37,000 --> 00:24:38,000 Oops. 484 00:24:39,000 --> 00:24:40,000 Clear the screen. 485 00:24:40,000 --> 00:24:40,000 Okay. 486 00:24:40,000 --> 00:24:44,000 Python llama 2.py. 487 00:24:44,000 --> 00:24:45,000 Okay. 488 00:24:48,000 --> 00:24:50,000 So it'll take some time again. 489 00:24:50,000 --> 00:24:56,000 Uh, as you all know, uh, some response time it will take with respect to the APIs that we have. 490 00:24:56,000 --> 00:25:01,000 Uh, and again, now we are hitting the specific API from the cloud, uh, from the AWS server itself, 491 00:25:01,000 --> 00:25:01,000 right. 492 00:25:01,000 --> 00:25:02,000 In AWS bedrock. 493 00:25:03,000 --> 00:25:06,000 So, so here is the entire text. 494 00:25:06,000 --> 00:25:12,000 Here you'll be able to see in the fair digital rain where data dot flows free. 495 00:25:12,000 --> 00:25:13,000 A wondrous art dot. 496 00:25:13,000 --> 00:25:15,000 RAR is called machine learning. 497 00:25:15,000 --> 00:25:19,000 You see this is a this is a science, a craft, a mystic spell. 498 00:25:19,000 --> 00:25:23,000 The dot enable machine to learn and tell with algorithm sharp and data set vast. 499 00:25:23,000 --> 00:25:27,000 Our computers do not don't gain wisdom and their insights do not. 500 00:25:27,000 --> 00:25:27,000 Last. 501 00:25:27,000 --> 00:25:30,000 And again based on the token size it will be charging. 502 00:25:30,000 --> 00:25:34,000 Okay, so all the information see you know whatever prompt you want to give over here. 503 00:25:34,000 --> 00:25:35,000 Right. 504 00:25:35,000 --> 00:25:39,000 Let's say I'll go ahead and write what is generative AI like. 505 00:25:39,000 --> 00:25:42,000 Um write a poem on generative AI. 506 00:25:42,000 --> 00:25:43,000 You'll be able to see this. 507 00:25:43,000 --> 00:25:45,000 You'll get some response over here. 508 00:25:45,000 --> 00:25:46,000 So I'll save this. 509 00:25:46,000 --> 00:25:49,000 Now I will go ahead and run it okay. 510 00:25:50,000 --> 00:25:52,000 Now similarly you can actually do it with cloudy too. 511 00:25:52,000 --> 00:25:54,000 You can do it with the stable diffusion. 512 00:25:54,000 --> 00:25:57,000 I'll also show you an example with respect to stable diffusion. 513 00:25:57,000 --> 00:26:01,000 And then you can also explore that uh as said uh now let's go ahead. 514 00:26:01,000 --> 00:26:04,000 Uh, you know, we will try to do it with the help of cloudy two. 515 00:26:04,000 --> 00:26:07,000 And again, uh, with respect to that also, we'll try to see okay. 516 00:26:07,000 --> 00:26:09,000 So this is one text generation. 517 00:26:09,000 --> 00:26:15,000 So here also you can see um, in the realm of code and circularity, a marvel of man's ingenuity, a 518 00:26:15,000 --> 00:26:17,000 creation that not rival the generative. 519 00:26:17,000 --> 00:26:21,000 I wonder to see so all the information are specifically coming over here. 520 00:26:21,000 --> 00:26:23,000 Now let's try another API. 521 00:26:23,000 --> 00:26:29,000 And then after this we will go ahead and try, uh, you know, the image generation with the help of 522 00:26:29,000 --> 00:26:30,000 stable diffusion. 523 00:26:30,000 --> 00:26:31,000 So cloudy. 524 00:26:31,000 --> 00:26:33,000 Uh, let's see where is cloudy. 525 00:26:33,000 --> 00:26:35,000 Cloudy content generation. 526 00:26:35,000 --> 00:26:36,000 Okay. 527 00:26:36,000 --> 00:26:38,000 So so here is my API. 528 00:26:38,000 --> 00:26:40,000 I will copy this. 529 00:26:40,000 --> 00:26:41,000 Um, um. 530 00:26:42,000 --> 00:26:45,000 It is always good to keep a test JSON like this. 531 00:26:45,000 --> 00:26:49,000 Okay, now almost each and everything is same okay I will tell talk about like what? 532 00:26:49,000 --> 00:26:50,000 All changes will basically happen. 533 00:26:50,000 --> 00:26:54,000 Okay, so llama two uh, let me just copy this. 534 00:26:54,000 --> 00:26:56,000 Four lines of code will be almost same. 535 00:26:56,000 --> 00:26:58,000 Then I have my payload. 536 00:26:58,000 --> 00:27:01,000 I will also copy this and then we'll change the payload. 537 00:27:01,000 --> 00:27:03,000 Okay, so we'll change the payload. 538 00:27:03,000 --> 00:27:07,000 So inside this payload you have prompt is equal to human. 539 00:27:07,000 --> 00:27:08,000 You are an expert social media generation. 540 00:27:08,000 --> 00:27:11,000 So whatever prompt basic prompt data right. 541 00:27:11,000 --> 00:27:16,000 So here I don't have to probably create this I don't want this okay. 542 00:27:16,000 --> 00:27:19,000 So this will basically be my prompt data done. 543 00:27:19,000 --> 00:27:21,000 Then let's go with the next parameter. 544 00:27:21,000 --> 00:27:24,000 So what are the next parameter. 545 00:27:24,000 --> 00:27:26,000 La la la la la la la. 546 00:27:26,000 --> 00:27:28,000 Baba baba baba. 547 00:27:29,000 --> 00:27:30,000 Come on guys. 548 00:27:30,000 --> 00:27:30,000 Till then. 549 00:27:30,000 --> 00:27:32,000 Hit like I'm working so much hard here. 550 00:27:32,000 --> 00:27:33,000 Come on. 551 00:27:34,000 --> 00:27:36,000 We we we will. 552 00:27:36,000 --> 00:27:36,000 Okay. 553 00:27:36,000 --> 00:27:37,000 Anthropic version is there. 554 00:27:37,000 --> 00:27:38,000 Stop sequence. 555 00:27:38,000 --> 00:27:39,000 Is there temperature? 556 00:27:39,000 --> 00:27:40,000 Uh, this is there. 557 00:27:40,000 --> 00:27:43,000 Max tokens is there say max token to sample is there? 558 00:27:43,000 --> 00:27:47,000 So let me just quickly go ahead and write what all things I require over here. 559 00:27:47,000 --> 00:27:48,000 So I will go ahead and write. 560 00:27:48,000 --> 00:27:49,000 Max. 561 00:27:49,000 --> 00:27:51,000 Uh, tokens. 562 00:27:51,000 --> 00:27:52,000 Okay. 563 00:27:52,000 --> 00:27:54,000 So this will be 5 to 12, uh, max tokens. 564 00:27:54,000 --> 00:27:55,000 Then I have my temperature. 565 00:27:55,000 --> 00:27:59,000 Then I have my the stop parameter looks something like this. 566 00:27:59,000 --> 00:27:59,000 Okay. 567 00:27:59,000 --> 00:28:02,000 Power point top P will be point eight. 568 00:28:02,000 --> 00:28:03,000 Okay. 569 00:28:03,000 --> 00:28:05,000 Something I'm putting some values. 570 00:28:05,000 --> 00:28:05,000 Temperature. 571 00:28:05,000 --> 00:28:07,000 Let's keep it 2.8. 572 00:28:07,000 --> 00:28:09,000 Let it be more creative okay. 573 00:28:09,000 --> 00:28:11,000 So this is basically my payload okay. 574 00:28:11,000 --> 00:28:12,000 Now everything almost looks same. 575 00:28:13,000 --> 00:28:16,000 Uh, you'll be able to see that I will uh mhm. 576 00:28:16,000 --> 00:28:16,000 Mhm. 577 00:28:17,000 --> 00:28:17,000 Okay. 578 00:28:17,000 --> 00:28:20,000 I think uh there is another one. 579 00:28:20,000 --> 00:28:21,000 Let's see. 580 00:28:21,000 --> 00:28:24,000 Content generation is also there. 581 00:28:24,000 --> 00:28:25,000 This is also there. 582 00:28:25,000 --> 00:28:26,000 Okay fine. 583 00:28:26,000 --> 00:28:27,000 No worries. 584 00:28:28,000 --> 00:28:28,000 Okay. 585 00:28:28,000 --> 00:28:31,000 So I will I think I've used another version. 586 00:28:31,000 --> 00:28:33,000 Let me use another one also which I have already tried. 587 00:28:33,000 --> 00:28:35,000 So it will be looking good okay. 588 00:28:35,000 --> 00:28:36,000 So cloudy two. 589 00:28:36,000 --> 00:28:40,000 So let's quickly go ahead and create my body. 590 00:28:40,000 --> 00:28:42,000 So here you have this one. 591 00:28:42,000 --> 00:28:46,000 So body JSON dumps payload model ID will be this one right. 592 00:28:46,000 --> 00:28:51,000 Bedrock this everything is there body model ID application and content type. 593 00:28:51,000 --> 00:28:56,000 So once we get the response this time when we get the response body okay. 594 00:28:56,000 --> 00:28:57,000 Like this. 595 00:28:57,000 --> 00:29:02,000 When we get this specific body now inside this body you will be able to see that they're in the case 596 00:29:02,000 --> 00:29:03,000 of llama two. 597 00:29:03,000 --> 00:29:05,000 Inside this body we have generation. 598 00:29:05,000 --> 00:29:10,000 But here inside this body we will be having some more steps okay. 599 00:29:10,000 --> 00:29:12,000 So that will be some more, uh, key value pairs. 600 00:29:12,000 --> 00:29:14,000 So here I've written it. 601 00:29:14,000 --> 00:29:17,000 So it will be nothing but completions of zero. 602 00:29:17,000 --> 00:29:19,000 Get data and then get text. 603 00:29:19,000 --> 00:29:23,000 So in completions there it is just like a list of key value pairs. 604 00:29:23,000 --> 00:29:28,000 So I will just take the first one and then probably display the data inside that whatever text is basically 605 00:29:28,000 --> 00:29:29,000 coming. 606 00:29:29,000 --> 00:29:34,000 And then we will go ahead and response underscore text okay. 607 00:29:34,000 --> 00:29:39,000 So now everything works fine I think I've used this model or other model. 608 00:29:39,000 --> 00:29:41,000 So this is version two okay. 609 00:29:41,000 --> 00:29:44,000 In cloudy let's say in cloudy. 610 00:29:44,000 --> 00:29:45,000 Some other version is also there. 611 00:29:46,000 --> 00:29:51,000 Um I don't know from where I found out, but it was available here only. 612 00:29:53,000 --> 00:29:54,000 Advanced Q&A. 613 00:29:54,000 --> 00:29:56,000 Yes, I used I used this one, I guess. 614 00:29:57,000 --> 00:29:58,000 Yeah. 615 00:29:58,000 --> 00:29:59,000 Advanced Q&A. 616 00:30:00,000 --> 00:30:02,000 I used this model ID, I guess. 617 00:30:03,000 --> 00:30:04,000 Okay, perfect. 618 00:30:04,000 --> 00:30:08,000 So once we do this, uh, let's go ahead and run it now. 619 00:30:08,000 --> 00:30:10,000 Now I'll go to my terminal. 620 00:30:11,000 --> 00:30:14,000 Now let's see whether each and every thing will run fine or not. 621 00:30:14,000 --> 00:30:16,000 Once I tried with llama two. 622 00:30:16,000 --> 00:30:17,000 Now I'm tried with cow two. 623 00:30:17,000 --> 00:30:18,000 Right? 624 00:30:18,000 --> 00:30:24,000 So I will write Python cloudy to cloudy.py. 625 00:30:24,000 --> 00:30:30,000 Okay, so here also you'll be able to see the print statement how this was quite fast right. 626 00:30:30,000 --> 00:30:35,000 So here you can see O gentle AI with your boundary power to think and create with no efforts. 627 00:30:35,000 --> 00:30:38,000 So you are specifically getting this specific output also. 628 00:30:38,000 --> 00:30:40,000 So this was with respect to cloudy. 629 00:30:40,000 --> 00:30:42,000 Again you can try anything that you want. 630 00:30:42,000 --> 00:30:43,000 Just take this JSON. 631 00:30:43,000 --> 00:30:44,000 Just try to play with it. 632 00:30:45,000 --> 00:30:49,000 Uh, let's say I take this information only and use this model. 633 00:30:49,000 --> 00:30:52,000 Let's see I will paste it over here. 634 00:30:52,000 --> 00:30:55,000 I will take this entire thing and paste this model. 635 00:30:56,000 --> 00:31:00,000 So here itself I'll change it. 636 00:31:03,000 --> 00:31:08,000 I think cloudy, I may get an error because I don't know whether we are able to manage it or not. 637 00:31:08,000 --> 00:31:09,000 So I sold Access denied. 638 00:31:09,000 --> 00:31:10,000 See? 639 00:31:10,000 --> 00:31:12,000 So this model is not available. 640 00:31:12,000 --> 00:31:13,000 The version one is available I guess. 641 00:31:13,000 --> 00:31:17,000 I think I, I explored this earlier find it somewhere. 642 00:31:17,000 --> 00:31:20,000 If I probably search this you will be able to find it I guess. 643 00:31:20,000 --> 00:31:21,000 Let's see. 644 00:31:22,000 --> 00:31:26,000 But anyhow, you try it out from your end I any I'll be giving you the code Like this. 645 00:31:26,000 --> 00:31:30,000 We are not able to search advanced Q and A and all. 646 00:31:30,000 --> 00:31:30,000 Okay, fine. 647 00:31:31,000 --> 00:31:36,000 Uh, let's start with, uh, I think, uh, c in US why that is not working. 648 00:31:36,000 --> 00:31:36,000 Right. 649 00:31:36,000 --> 00:31:41,000 Because if I go ahead and see the model access, the cloudy model is not available. 650 00:31:41,000 --> 00:31:44,000 C cloudy instant is not available. 651 00:31:44,000 --> 00:31:46,000 So that is the reason that version one was available. 652 00:31:46,000 --> 00:31:47,000 And I was still exploring things. 653 00:31:47,000 --> 00:31:48,000 Right. 654 00:31:48,000 --> 00:31:52,000 And I probably got in that, But no worries, I will now go ahead with stable diffusion. 655 00:31:52,000 --> 00:31:57,000 So let's go to the base model, see version 2.1. 656 00:31:57,000 --> 00:31:59,000 This is all models are specifically there. 657 00:31:59,000 --> 00:32:02,000 If I go to examples uh let's see with respect to stable diffusion. 658 00:32:02,000 --> 00:32:04,000 So 1.2 is there 0.8 is there. 659 00:32:04,000 --> 00:32:06,000 We'll go with the recent one. 660 00:32:06,000 --> 00:32:09,000 And again I will go ahead and copy this okay. 661 00:32:10,000 --> 00:32:14,000 So this will basically be my next text JSON okay. 662 00:32:14,000 --> 00:32:16,000 So I will keep all these things over here itself. 663 00:32:16,000 --> 00:32:19,000 Now let me quickly create one more file. 664 00:32:19,000 --> 00:32:23,000 Save this one more file stable diffusion. 665 00:32:23,000 --> 00:32:28,000 Now here we will try to create an image in AWS itself from AWS dot Pi. 666 00:32:28,000 --> 00:32:30,000 Okay now for stable diffusion. 667 00:32:30,000 --> 00:32:32,000 Also almost all the things will be same. 668 00:32:32,000 --> 00:32:37,000 I've already created the code, but here what I will do, I will just show you what all things I will 669 00:32:37,000 --> 00:32:37,000 do. 670 00:32:37,000 --> 00:32:40,000 See, stable diffusion basically means I have written a prompt data. 671 00:32:40,000 --> 00:32:41,000 See? 672 00:32:41,000 --> 00:32:43,000 Provided me a 4k HD image of beach. 673 00:32:43,000 --> 00:32:45,000 Also use blue sky, rainy season and cinematic display. 674 00:32:45,000 --> 00:32:52,000 Okay, now with respect to this text JSON here you can see body has something called as text prompt. 675 00:32:52,000 --> 00:32:54,000 And then you have text information. 676 00:32:54,000 --> 00:32:57,000 And with respect to the text information this is basically my prompt. 677 00:32:57,000 --> 00:33:01,000 Along with this you have parameters like weight cfg, scale seed, this, this, this is there. 678 00:33:01,000 --> 00:33:05,000 So what I do first of all I will set in that format only. 679 00:33:05,000 --> 00:33:09,000 So first of all I create my prompt template where I have my text my prompt data with weight. 680 00:33:09,000 --> 00:33:09,000 Okay. 681 00:33:09,000 --> 00:33:12,000 So this becomes my entire this information. 682 00:33:12,000 --> 00:33:13,000 Understand this thing right. 683 00:33:13,000 --> 00:33:14,000 You just need to set it. 684 00:33:14,000 --> 00:33:15,000 That's that's it. 685 00:33:15,000 --> 00:33:17,000 Text is equal to this information with weight. 686 00:33:17,000 --> 00:33:20,000 So I've kept it in the form of list with key value pairs. 687 00:33:20,000 --> 00:33:23,000 So that is what I am getting it over here inside this prompt template. 688 00:33:23,000 --> 00:33:27,000 Now I have used butter client I've used the service name runtime. 689 00:33:27,000 --> 00:33:29,000 Now remaining is all my payload. 690 00:33:29,000 --> 00:33:31,000 So inside this whatever payload you require. 691 00:33:31,000 --> 00:33:33,000 See first of all I require text prompt. 692 00:33:33,000 --> 00:33:37,000 The second thing that I require is seed steps width height right. 693 00:33:37,000 --> 00:33:38,000 Width height. 694 00:33:38,000 --> 00:33:39,000 So let me do one thing. 695 00:33:39,000 --> 00:33:42,000 I will change this to 1024. 696 00:33:44,000 --> 00:33:45,000 I will also change it to 1024. 697 00:33:45,000 --> 00:33:47,000 So these are all the parameters I've set. 698 00:33:47,000 --> 00:33:49,000 Step size is nothing but 50. 699 00:33:49,000 --> 00:33:50,000 Width is 1024. 700 00:33:50,000 --> 00:33:51,000 Height is 1024. 701 00:33:51,000 --> 00:33:53,000 So this becomes my entire payload. 702 00:33:53,000 --> 00:33:54,000 I have converted this into JSON. 703 00:33:54,000 --> 00:33:59,000 I've used this model ID because this is the same model id I'm able to get it over here right. 704 00:33:59,000 --> 00:34:02,000 Stability dot stable diffusion X1V1. 705 00:34:02,000 --> 00:34:03,000 It's very simple. 706 00:34:03,000 --> 00:34:05,000 Once you do one right everything will be able to use it. 707 00:34:05,000 --> 00:34:06,000 So here I've used that. 708 00:34:06,000 --> 00:34:10,000 Then I have invoked this model with content type model ID body everything. 709 00:34:10,000 --> 00:34:12,000 Once I get the response. 710 00:34:12,000 --> 00:34:15,000 Now see what response you will get in this format right? 711 00:34:15,000 --> 00:34:20,000 When you probably see the response body there, you'll be getting key value pairs with something called 712 00:34:20,000 --> 00:34:21,000 as artifacts. 713 00:34:21,000 --> 00:34:26,000 Inside that you will find a base 64 encoded response of the images. 714 00:34:26,000 --> 00:34:29,000 So what we did is that we took this particular artifact of zero. 715 00:34:29,000 --> 00:34:31,000 We took that base 64. 716 00:34:31,000 --> 00:34:33,000 We encoded it with UTF eight. 717 00:34:33,000 --> 00:34:37,000 And this is how you read any encoded image right from that encoded image. 718 00:34:37,000 --> 00:34:38,000 We have converted that into bytes. 719 00:34:38,000 --> 00:34:42,000 And then we have saved that particular image to some output directory. 720 00:34:42,000 --> 00:34:44,000 So output directory output folder will get created. 721 00:34:44,000 --> 00:34:47,000 And I am creating one image over here. 722 00:34:47,000 --> 00:34:47,000 Right. 723 00:34:47,000 --> 00:34:51,000 So uh that is what we are specifically doing with respect to this. 724 00:34:51,000 --> 00:34:51,000 Right. 725 00:34:51,000 --> 00:34:55,000 So uh file name will be nothing but inside that particular output directory. 726 00:34:55,000 --> 00:34:57,000 And then we can open that specific image if you want. 727 00:34:57,000 --> 00:34:58,000 Okay. 728 00:34:58,000 --> 00:34:59,000 We are writing that image. 729 00:34:59,000 --> 00:34:59,000 Sorry. 730 00:34:59,000 --> 00:35:01,000 We are writing that image inside this particular PNG. 731 00:35:01,000 --> 00:35:02,000 That is same. 732 00:35:02,000 --> 00:35:07,000 See whatever things we did over here, only the thing that is changing how you converting that bytes 733 00:35:07,000 --> 00:35:10,000 information that we are getting into an image and saving that in our folder. 734 00:35:10,000 --> 00:35:11,000 Okay. 735 00:35:11,000 --> 00:35:14,000 So let me quickly execute and run it and show it to you. 736 00:35:14,000 --> 00:35:17,000 And then you will be able to understand okay. 737 00:35:18,000 --> 00:35:23,000 So here I will be showing you Python stable diffusion dot pi. 738 00:35:23,000 --> 00:35:28,000 So once I execute it here you know okay request steps 400 generation error. 739 00:35:28,000 --> 00:35:29,000 What is this error? 740 00:35:29,000 --> 00:35:30,000 Let's see. 741 00:35:31,000 --> 00:35:34,000 Uh, so okay, there is an error. 742 00:35:34,000 --> 00:35:36,000 Let's see what is the error. 743 00:35:36,000 --> 00:35:39,000 So bedrock invoke invoke model. 744 00:35:39,000 --> 00:35:40,000 Let me see. 745 00:35:46,000 --> 00:35:49,000 I think there was a small minor mistake. 746 00:35:49,000 --> 00:35:50,000 Which one version. 747 00:35:50,000 --> 00:35:52,000 This this this slight entry. 748 00:35:52,000 --> 00:35:52,000 Okay. 749 00:35:52,000 --> 00:35:54,000 And also let me see here. 750 00:35:54,000 --> 00:35:54,000 Yeah. 751 00:35:58,000 --> 00:35:59,000 So now it is working. 752 00:35:59,000 --> 00:36:02,000 I did some minor mistake over there which I have fixed it. 753 00:36:02,000 --> 00:36:02,000 Okay. 754 00:36:04,000 --> 00:36:06,000 So this is the byte information that we are getting. 755 00:36:06,000 --> 00:36:07,000 See the image bytes. 756 00:36:07,000 --> 00:36:11,000 Now if I see my output folder this is my image okay. 757 00:36:11,000 --> 00:36:19,000 So if I go ahead and see find in folder okay sorry I will go ahead and preview in folder reveal in File 758 00:36:19,000 --> 00:36:19,000 Explorer. 759 00:36:19,000 --> 00:36:22,000 So if I go ahead and create this image see this image. 760 00:36:22,000 --> 00:36:24,000 This is how the image is basically created. 761 00:36:24,000 --> 00:36:25,000 And it's look good right. 762 00:36:25,000 --> 00:36:27,000 But what is the prompt I've used over here? 763 00:36:27,000 --> 00:36:35,000 If you see uh, with respect to this, uh, my stable diffusion dot p y here, we have basically created 764 00:36:35,000 --> 00:36:36,000 something like this. 765 00:36:36,000 --> 00:36:38,000 Provide me a 4k HD image of a beach. 766 00:36:38,000 --> 00:36:40,000 So I hope you got an idea with respect to this. 767 00:36:40,000 --> 00:36:44,000 Now in the upcoming videos, what I'll do is that I'll create some amusing projects by using this specific 768 00:36:44,000 --> 00:36:45,000 APIs. 769 00:36:45,000 --> 00:36:48,000 But in short, once I deploy these projects, I can deploy it any way I want, right? 770 00:36:49,000 --> 00:36:51,000 Hugging face spaces anywhere itself. 771 00:36:51,000 --> 00:36:56,000 So my suggestion would be that try as many as you can see different different like character role play. 772 00:36:56,000 --> 00:36:59,000 What is the input prompt you really need to give based on that? 773 00:36:59,000 --> 00:37:02,000 Try to create use cases which will be amazing for you. 774 00:37:02,000 --> 00:37:02,000 Right? 775 00:37:02,000 --> 00:37:04,000 So this was it from my side. 776 00:37:04,000 --> 00:37:04,000 I hope you liked this particular.