WEBVTT Kind: captions Language: en 00:00:00.240 --> 00:00:02.470 align:start position:0% So,<00:00:00.480> looks<00:00:00.719> like<00:00:00.880> we<00:00:01.120> just<00:00:01.360> got<00:00:01.600> flashed<00:00:02.080> by 00:00:02.470 --> 00:00:02.480 align:start position:0% So, looks like we just got flashed by 00:00:02.480 --> 00:00:04.630 align:start position:0% So, looks like we just got flashed by Google<00:00:02.800> DeepMind<00:00:03.274> [music]<00:00:03.600> once<00:00:03.919> again 00:00:04.630 --> 00:00:04.640 align:start position:0% Google DeepMind [music] once again 00:00:04.640 --> 00:00:06.869 align:start position:0% Google DeepMind [music] once again because<00:00:04.960> they<00:00:05.200> just<00:00:05.359> dropped<00:00:05.680> Gemini<00:00:06.160> 3.7 00:00:06.869 --> 00:00:06.879 align:start position:0% because they just dropped Gemini 3.7 00:00:06.879 --> 00:00:09.430 align:start position:0% because they just dropped Gemini 3.7 Flash<00:00:07.279> today.<00:00:07.919> Yes,<00:00:08.240> their<00:00:08.559> newest<00:00:08.960> model, 00:00:09.430 --> 00:00:09.440 align:start position:0% Flash today. Yes, their newest model, 00:00:09.440 --> 00:00:11.669 align:start position:0% Flash today. Yes, their newest model, which<00:00:09.679> is<00:00:09.920> once<00:00:10.160> again<00:00:10.400> a<00:00:10.639> Flash<00:00:11.040> model,<00:00:11.360> is 00:00:11.669 --> 00:00:11.679 align:start position:0% which is once again a Flash model, is 00:00:11.679 --> 00:00:14.150 align:start position:0% which is once again a Flash model, is here.<00:00:12.240> And<00:00:12.400> this<00:00:12.639> is<00:00:12.800> only<00:00:13.120> after<00:00:13.519> 3<00:00:13.759> weeks 00:00:14.150 --> 00:00:14.160 align:start position:0% here. And this is only after 3 weeks 00:00:14.160 --> 00:00:17.269 align:start position:0% here. And this is only after 3 weeks since<00:00:14.559> Gemini<00:00:15.120> 3.6<00:00:15.920> Flash.<00:00:16.720> And<00:00:16.880> what<00:00:17.119> they 00:00:17.269 --> 00:00:17.279 align:start position:0% since Gemini 3.6 Flash. And what they 00:00:17.279 --> 00:00:19.990 align:start position:0% since Gemini 3.6 Flash. And what they are<00:00:17.520> saying<00:00:17.760> is<00:00:17.920> that<00:00:18.240> Gemini<00:00:18.720> 3.7<00:00:19.439> Flash<00:00:19.760> is 00:00:19.990 --> 00:00:20.000 align:start position:0% are saying is that Gemini 3.7 Flash is 00:00:20.000 --> 00:00:22.790 align:start position:0% are saying is that Gemini 3.7 Flash is their<00:00:20.240> most<00:00:20.560> intelligent<00:00:21.359> workhorse<00:00:22.000> model. 00:00:22.790 --> 00:00:22.800 align:start position:0% their most intelligent workhorse model. 00:00:22.800 --> 00:00:25.349 align:start position:0% their most intelligent workhorse model. And<00:00:22.960> this<00:00:23.199> is<00:00:23.359> coming<00:00:23.680> 3<00:00:23.920> weeks<00:00:24.160> after<00:00:24.640> 3.6<00:00:25.119> six 00:00:25.349 --> 00:00:25.359 align:start position:0% And this is coming 3 weeks after 3.6 six 00:00:25.359 --> 00:00:27.830 align:start position:0% And this is coming 3 weeks after 3.6 six flash<00:00:25.760> because<00:00:26.160> of<00:00:26.400> developer<00:00:26.960> feedback<00:00:27.519> and 00:00:27.830 --> 00:00:27.840 align:start position:0% flash because of developer feedback and 00:00:27.840 --> 00:00:30.470 align:start position:0% flash because of developer feedback and algorithmic<00:00:28.560> innovations.<00:00:29.920> And<00:00:30.080> that<00:00:30.240> is<00:00:30.320> why 00:00:30.470 --> 00:00:30.480 align:start position:0% algorithmic innovations. And that is why 00:00:30.480 --> 00:00:32.470 align:start position:0% algorithmic innovations. And that is why they're<00:00:30.720> saying<00:00:30.800> the<00:00:30.960> Gemini<00:00:31.359> 3.7<00:00:32.079> Flash 00:00:32.470 --> 00:00:32.480 align:start position:0% they're saying the Gemini 3.7 Flash 00:00:32.480 --> 00:00:34.310 align:start position:0% they're saying the Gemini 3.7 Flash model<00:00:32.880> is,<00:00:33.200> you<00:00:33.280> know,<00:00:33.440> out<00:00:33.680> with<00:00:33.920> so<00:00:34.160> much 00:00:34.310 --> 00:00:34.320 align:start position:0% model is, you know, out with so much 00:00:34.320 --> 00:00:36.150 align:start position:0% model is, you know, out with so much improvement<00:00:35.200> cuz<00:00:35.440> if<00:00:35.600> we<00:00:35.680> take<00:00:35.760> a<00:00:35.840> look<00:00:35.920> at<00:00:36.079> the 00:00:36.150 --> 00:00:36.160 align:start position:0% improvement cuz if we take a look at the 00:00:36.160 --> 00:00:38.389 align:start position:0% improvement cuz if we take a look at the benchmarks,<00:00:36.800> this<00:00:36.960> model<00:00:37.360> is<00:00:38.000> way<00:00:38.239> better 00:00:38.389 --> 00:00:38.399 align:start position:0% benchmarks, this model is way better 00:00:38.399 --> 00:00:41.350 align:start position:0% benchmarks, this model is way better than<00:00:38.559> the<00:00:38.719> Gemini<00:00:39.200> 3.6<00:00:39.840> Flash.<00:00:40.719> But<00:00:40.879> I<00:00:41.120> think 00:00:41.350 --> 00:00:41.360 align:start position:0% than the Gemini 3.6 Flash. But I think 00:00:41.360 --> 00:00:43.430 align:start position:0% than the Gemini 3.6 Flash. But I think this<00:00:41.600> is<00:00:41.840> also<00:00:42.160> because<00:00:42.399> we<00:00:42.640> know<00:00:42.879> internally 00:00:43.430 --> 00:00:43.440 align:start position:0% this is also because we know internally 00:00:43.440 --> 00:00:44.869 align:start position:0% this is also because we know internally that<00:00:43.680> they're<00:00:44.000> planning<00:00:44.239> on<00:00:44.480> cancelelling 00:00:44.869 --> 00:00:44.879 align:start position:0% that they're planning on cancelelling 00:00:44.879 --> 00:00:48.150 align:start position:0% that they're planning on cancelelling Gemini<00:00:45.360> 3.5<00:00:46.160> Pro<00:00:46.559> completely<00:00:47.440> and<00:00:47.760> working<00:00:48.000> on 00:00:48.150 --> 00:00:48.160 align:start position:0% Gemini 3.5 Pro completely and working on 00:00:48.160 --> 00:00:50.470 align:start position:0% Gemini 3.5 Pro completely and working on Gemini<00:00:48.719> 4<00:00:48.960> at<00:00:49.200> the<00:00:49.360> moment.<00:00:49.840> So<00:00:50.000> maybe<00:00:50.239> the 00:00:50.470 --> 00:00:50.480 align:start position:0% Gemini 4 at the moment. So maybe the 00:00:50.480 --> 00:00:51.910 align:start position:0% Gemini 4 at the moment. So maybe the improvements<00:00:50.879> they<00:00:51.120> made<00:00:51.280> with<00:00:51.440> the<00:00:51.600> Pro 00:00:51.910 --> 00:00:51.920 align:start position:0% improvements they made with the Pro 00:00:51.920 --> 00:00:53.510 align:start position:0% improvements they made with the Pro model<00:00:52.320> that<00:00:52.559> they're<00:00:52.800> not<00:00:53.039> going<00:00:53.199> to<00:00:53.360> be 00:00:53.510 --> 00:00:53.520 align:start position:0% model that they're not going to be 00:00:53.520 --> 00:00:55.830 align:start position:0% model that they're not going to be dropping<00:00:53.920> anymore.<00:00:54.640> Maybe<00:00:54.879> they<00:00:55.120> repackaged 00:00:55.830 --> 00:00:55.840 align:start position:0% dropping anymore. Maybe they repackaged 00:00:55.840 --> 00:00:58.310 align:start position:0% dropping anymore. Maybe they repackaged it<00:00:56.000> as<00:00:56.160> a<00:00:56.320> flash<00:00:56.719> model<00:00:57.039> because<00:00:57.680> this<00:00:58.000> model 00:00:58.310 --> 00:00:58.320 align:start position:0% it as a flash model because this model 00:00:58.320 --> 00:01:01.189 align:start position:0% it as a flash model because this model is<00:00:58.559> better<00:00:58.800> than<00:00:59.199> 3.6<00:01:00.000> and<00:01:00.239> if<00:01:00.559> the<00:01:00.800> timeline 00:01:01.189 --> 00:01:01.199 align:start position:0% is better than 3.6 and if the timeline 00:01:01.199 --> 00:01:03.110 align:start position:0% is better than 3.6 and if the timeline is<00:01:01.440> three<00:01:01.680> weeks<00:01:02.000> plus<00:01:02.239> they're<00:01:02.559> planning<00:01:02.879> on 00:01:03.110 --> 00:01:03.120 align:start position:0% is three weeks plus they're planning on 00:01:03.120 --> 00:01:05.509 align:start position:0% is three weeks plus they're planning on cancelling<00:01:03.680> the<00:01:03.840> 3.5<00:01:04.479> Pro,<00:01:05.040> it's<00:01:05.360> quite 00:01:05.509 --> 00:01:05.519 align:start position:0% cancelling the 3.5 Pro, it's quite 00:01:05.519 --> 00:01:07.590 align:start position:0% cancelling the 3.5 Pro, it's quite likely<00:01:05.840> that<00:01:06.080> is<00:01:06.240> the<00:01:06.479> case.<00:01:07.119> We'll<00:01:07.360> just 00:01:07.590 --> 00:01:07.600 align:start position:0% likely that is the case. We'll just 00:01:07.600 --> 00:01:09.109 align:start position:0% likely that is the case. We'll just ignore<00:01:07.760> that<00:01:07.920> and<00:01:08.080> put<00:01:08.159> that<00:01:08.320> into<00:01:08.560> the<00:01:08.799> side. 00:01:09.109 --> 00:01:09.119 align:start position:0% ignore that and put that into the side. 00:01:09.119 --> 00:01:10.469 align:start position:0% ignore that and put that into the side. We<00:01:09.280> don't<00:01:09.439> know<00:01:09.520> when<00:01:09.760> we're<00:01:09.920> getting<00:01:10.080> a<00:01:10.320> new 00:01:10.469 --> 00:01:10.479 align:start position:0% We don't know when we're getting a new 00:01:10.479 --> 00:01:12.070 align:start position:0% We don't know when we're getting a new Pro<00:01:10.720> model<00:01:10.960> from<00:01:11.200> the<00:01:11.360> Google<00:01:11.600> Deep<00:01:11.840> Mind 00:01:12.070 --> 00:01:12.080 align:start position:0% Pro model from the Google Deep Mind 00:01:12.080 --> 00:01:14.390 align:start position:0% Pro model from the Google Deep Mind team,<00:01:12.560> but<00:01:12.799> this<00:01:13.040> model<00:01:13.520> across<00:01:13.920> many<00:01:14.159> of<00:01:14.240> the 00:01:14.390 --> 00:01:14.400 align:start position:0% team, but this model across many of the 00:01:14.400 --> 00:01:17.109 align:start position:0% team, but this model across many of the benchmarks<00:01:15.040> we<00:01:15.200> will<00:01:15.439> see<00:01:16.000> that<00:01:16.320> yes,<00:01:16.640> it<00:01:16.880> is 00:01:17.109 --> 00:01:17.119 align:start position:0% benchmarks we will see that yes, it is 00:01:17.119 --> 00:01:19.910 align:start position:0% benchmarks we will see that yes, it is stronger<00:01:17.439> than<00:01:17.680> the<00:01:17.920> 3.6<00:01:18.720> Flash<00:01:19.119> model.<00:01:19.759> For 00:01:19.910 --> 00:01:19.920 align:start position:0% stronger than the 3.6 Flash model. For 00:01:19.920 --> 00:01:21.830 align:start position:0% stronger than the 3.6 Flash model. For example,<00:01:20.320> when<00:01:20.479> it<00:01:20.640> comes<00:01:20.799> to<00:01:20.960> code<00:01:21.280> quality, 00:01:21.830 --> 00:01:21.840 align:start position:0% example, when it comes to code quality, 00:01:21.840 --> 00:01:24.149 align:start position:0% example, when it comes to code quality, production<00:01:22.320> code<00:01:22.640> quality,<00:01:23.280> the<00:01:23.520> model<00:01:23.840> is 00:01:24.149 --> 00:01:24.159 align:start position:0% production code quality, the model is 00:01:24.159 --> 00:01:26.149 align:start position:0% production code quality, the model is producing<00:01:24.640> 43.6% 00:01:26.149 --> 00:01:26.159 align:start position:0% producing 43.6% 00:01:26.159 --> 00:01:29.670 align:start position:0% producing 43.6% versus<00:01:26.640> the<00:01:27.200> flash<00:01:27.600> model<00:01:27.920> 34.4%. 00:01:29.670 --> 00:01:29.680 align:start position:0% versus the flash model 34.4%. 00:01:29.680 --> 00:01:33.510 align:start position:0% versus the flash model 34.4%. And<00:01:29.840> then<00:01:30.080> Sonet<00:01:30.640> 5<00:01:30.960> is<00:01:31.439> at<00:01:31.759> 42.7% 00:01:33.510 --> 00:01:33.520 align:start position:0% And then Sonet 5 is at 42.7% 00:01:33.520 --> 00:01:36.789 align:start position:0% And then Sonet 5 is at 42.7% and<00:01:33.759> the<00:01:33.920> Terra<00:01:34.320> model<00:01:34.640> is<00:01:34.880> at<00:01:35.280> 41.3%. 00:01:36.789 --> 00:01:36.799 align:start position:0% and the Terra model is at 41.3%. 00:01:36.799 --> 00:01:38.789 align:start position:0% and the Terra model is at 41.3%. So,<00:01:37.040> one<00:01:37.280> thing<00:01:37.439> to<00:01:37.680> remember,<00:01:38.240> since<00:01:38.479> this<00:01:38.720> is 00:01:38.789 --> 00:01:38.799 align:start position:0% So, one thing to remember, since this is 00:01:38.799 --> 00:01:40.230 align:start position:0% So, one thing to remember, since this is a<00:01:38.960> flash<00:01:39.360> model,<00:01:39.680> you're<00:01:39.840> not<00:01:40.000> going<00:01:40.000> to<00:01:40.079> see 00:01:40.230 --> 00:01:40.240 align:start position:0% a flash model, you're not going to see 00:01:40.240 --> 00:01:42.149 align:start position:0% a flash model, you're not going to see them<00:01:40.400> compare<00:01:40.720> this<00:01:40.880> to<00:01:41.040> Opus<00:01:41.439> 5<00:01:41.680> or<00:01:41.840> any<00:01:42.000> of 00:01:42.149 --> 00:01:42.159 align:start position:0% them compare this to Opus 5 or any of 00:01:42.159 --> 00:01:44.230 align:start position:0% them compare this to Opus 5 or any of the<00:01:42.320> other<00:01:42.640> stronger<00:01:43.119> models.<00:01:43.759> The<00:01:44.000> reason 00:01:44.230 --> 00:01:44.240 align:start position:0% the other stronger models. The reason 00:01:44.240 --> 00:01:46.230 align:start position:0% the other stronger models. The reason being<00:01:44.720> cuz<00:01:44.960> this<00:01:45.119> is<00:01:45.360> not<00:01:45.520> their<00:01:45.840> strongest 00:01:46.230 --> 00:01:46.240 align:start position:0% being cuz this is not their strongest 00:01:46.240 --> 00:01:49.109 align:start position:0% being cuz this is not their strongest tier.<00:01:46.960> and<00:01:47.200> the<00:01:47.360> 3.1<00:01:48.079> Pro<00:01:48.399> model<00:01:48.640> when<00:01:48.880> they 00:01:49.109 --> 00:01:49.119 align:start position:0% tier. and the 3.1 Pro model when they 00:01:49.119 --> 00:01:51.590 align:start position:0% tier. and the 3.1 Pro model when they finally<00:01:49.439> decide<00:01:49.759> to<00:01:50.079> upgrade<00:01:50.399> it<00:01:50.640> to<00:01:50.799> Gemini<00:01:51.360> 4 00:01:51.590 --> 00:01:51.600 align:start position:0% finally decide to upgrade it to Gemini 4 00:01:51.600 --> 00:01:54.469 align:start position:0% finally decide to upgrade it to Gemini 4 Pro,<00:01:52.960> then<00:01:53.200> we'll<00:01:53.360> see<00:01:53.520> it<00:01:53.680> being<00:01:53.920> compared<00:01:54.240> to 00:01:54.469 --> 00:01:54.479 align:start position:0% Pro, then we'll see it being compared to 00:01:54.479 --> 00:01:57.910 align:start position:0% Pro, then we'll see it being compared to Opus<00:01:54.880> 5<00:01:55.040> or<00:01:55.280> GPT<00:01:55.840> 5.6<00:01:56.399> Soul.<00:01:57.040> But<00:01:57.200> anyways,<00:01:57.680> we 00:01:57.910 --> 00:01:57.920 align:start position:0% Opus 5 or GPT 5.6 Soul. But anyways, we 00:01:57.920 --> 00:01:59.270 align:start position:0% Opus 5 or GPT 5.6 Soul. But anyways, we can<00:01:58.079> see<00:01:58.159> that<00:01:58.399> this<00:01:58.640> model<00:01:58.880> is<00:01:59.040> an 00:01:59.270 --> 00:01:59.280 align:start position:0% can see that this model is an 00:01:59.280 --> 00:02:01.510 align:start position:0% can see that this model is an improvement<00:01:59.600> from<00:01:59.920> 3.6<00:02:00.640> Flash<00:02:01.040> and<00:02:01.280> I<00:02:01.439> would 00:02:01.510 --> 00:02:01.520 align:start position:0% improvement from 3.6 Flash and I would 00:02:01.520 --> 00:02:04.550 align:start position:0% improvement from 3.6 Flash and I would say<00:02:01.680> that's<00:02:02.079> basically<00:02:03.280> the<00:02:03.680> biggest<00:02:04.159> result 00:02:04.550 --> 00:02:04.560 align:start position:0% say that's basically the biggest result 00:02:04.560 --> 00:02:06.469 align:start position:0% say that's basically the biggest result or<00:02:04.880> the<00:02:05.040> biggest<00:02:05.360> update<00:02:05.680> that<00:02:05.920> we<00:02:06.079> saw<00:02:06.240> with 00:02:06.469 --> 00:02:06.479 align:start position:0% or the biggest update that we saw with 00:02:06.479 --> 00:02:08.550 align:start position:0% or the biggest update that we saw with the<00:02:06.640> new<00:02:06.799> model<00:02:07.439> because<00:02:07.680> it<00:02:08.000> does<00:02:08.160> not<00:02:08.319> really 00:02:08.550 --> 00:02:08.560 align:start position:0% the new model because it does not really 00:02:08.560 --> 00:02:10.710 align:start position:0% the new model because it does not really like<00:02:08.879> change<00:02:09.119> up<00:02:09.440> things<00:02:09.679> a<00:02:09.920> lot<00:02:10.160> is<00:02:10.479> still 00:02:10.710 --> 00:02:10.720 align:start position:0% like change up things a lot is still 00:02:10.720 --> 00:02:12.550 align:start position:0% like change up things a lot is still like<00:02:10.959> not<00:02:11.200> becoming<00:02:11.599> the<00:02:11.840> number<00:02:12.000> one<00:02:12.319> model 00:02:12.550 --> 00:02:12.560 align:start position:0% like not becoming the number one model 00:02:12.560 --> 00:02:15.030 align:start position:0% like not becoming the number one model or<00:02:12.800> the<00:02:12.959> number<00:02:13.200> one<00:02:13.440> flash<00:02:13.840> model.<00:02:14.480> Even<00:02:14.800> on 00:02:15.030 --> 00:02:15.040 align:start position:0% or the number one flash model. Even on 00:02:15.040 --> 00:02:17.190 align:start position:0% or the number one flash model. Even on some<00:02:15.280> benchmarks,<00:02:15.920> Deepseek<00:02:16.560> version<00:02:16.879> 4 00:02:17.190 --> 00:02:17.200 align:start position:0% some benchmarks, Deepseek version 4 00:02:17.200 --> 00:02:19.910 align:start position:0% some benchmarks, Deepseek version 4 flash<00:02:17.680> is<00:02:18.000> actually<00:02:18.800> cheaper<00:02:19.280> and<00:02:19.680> more 00:02:19.910 --> 00:02:19.920 align:start position:0% flash is actually cheaper and more 00:02:19.920 --> 00:02:22.150 align:start position:0% flash is actually cheaper and more intelligent<00:02:20.480> than<00:02:20.720> this<00:02:21.040> model.<00:02:21.760> But<00:02:21.920> let's 00:02:22.150 --> 00:02:22.160 align:start position:0% intelligent than this model. But let's 00:02:22.160 --> 00:02:23.670 align:start position:0% intelligent than this model. But let's just<00:02:22.319> take<00:02:22.480> a<00:02:22.560> look<00:02:22.720> at<00:02:22.800> the<00:02:22.959> benchmarks<00:02:23.440> that 00:02:23.670 --> 00:02:23.680 align:start position:0% just take a look at the benchmarks that 00:02:23.680 --> 00:02:25.830 align:start position:0% just take a look at the benchmarks that they<00:02:23.840> have<00:02:24.080> published.<00:02:24.800> On<00:02:25.040> Long<00:02:25.360> Horizon 00:02:25.830 --> 00:02:25.840 align:start position:0% they have published. On Long Horizon 00:02:25.840 --> 00:02:28.070 align:start position:0% they have published. On Long Horizon software<00:02:26.160> engineering,<00:02:27.040> this<00:02:27.280> model<00:02:27.599> is<00:02:27.840> a 00:02:28.070 --> 00:02:28.080 align:start position:0% software engineering, this model is a 00:02:28.080 --> 00:02:30.949 align:start position:0% software engineering, this model is a little<00:02:28.160> bit<00:02:28.319> behind<00:02:28.800> GPT<00:02:29.360> 5.6<00:02:30.080> Terra,<00:02:30.640> which 00:02:30.949 --> 00:02:30.959 align:start position:0% little bit behind GPT 5.6 Terra, which 00:02:30.959 --> 00:02:32.949 align:start position:0% little bit behind GPT 5.6 Terra, which sits<00:02:31.200> at<00:02:31.440> 69.6% 00:02:32.949 --> 00:02:32.959 align:start position:0% sits at 69.6% 00:02:32.959 --> 00:02:36.390 align:start position:0% sits at 69.6% and<00:02:33.120> then<00:02:33.440> 65.3%<00:02:34.720> is<00:02:34.959> the<00:02:35.120> Flash<00:02:35.440> model.<00:02:36.160> The 00:02:36.390 --> 00:02:36.400 align:start position:0% and then 65.3% is the Flash model. The 00:02:36.400 --> 00:02:40.790 align:start position:0% and then 65.3% is the Flash model. The older<00:02:36.800> Flash<00:02:37.120> model<00:02:37.519> 3.6<00:02:38.560> sits<00:02:38.800> at<00:02:39.040> 48.6%. 00:02:40.790 --> 00:02:40.800 align:start position:0% older Flash model 3.6 sits at 48.6%. 00:02:40.800 --> 00:02:43.990 align:start position:0% older Flash model 3.6 sits at 48.6%. Sonic<00:02:41.200> 5<00:02:41.440> sits<00:02:41.760> at<00:02:41.920> 53.8%<00:02:42.720> 8%.<00:02:43.360> And<00:02:43.519> the<00:02:43.760> new 00:02:43.990 --> 00:02:44.000 align:start position:0% Sonic 5 sits at 53.8% 8%. And the new 00:02:44.000 --> 00:02:46.150 align:start position:0% Sonic 5 sits at 53.8% 8%. And the new player<00:02:44.319> that<00:02:44.560> is<00:02:44.800> finally<00:02:45.120> being<00:02:45.360> included<00:02:45.840> on 00:02:46.150 --> 00:02:46.160 align:start position:0% player that is finally being included on 00:02:46.160 --> 00:02:47.990 align:start position:0% player that is finally being included on benchmarks,<00:02:47.120> which<00:02:47.360> even<00:02:47.599> the<00:02:47.760> Google 00:02:47.990 --> 00:02:48.000 align:start position:0% benchmarks, which even the Google 00:02:48.000 --> 00:02:49.910 align:start position:0% benchmarks, which even the Google DeepMind<00:02:48.480> team<00:02:48.720> is<00:02:48.959> considering<00:02:49.360> with<00:02:49.599> their 00:02:49.910 --> 00:02:49.920 align:start position:0% DeepMind team is considering with their 00:02:49.920 --> 00:02:52.470 align:start position:0% DeepMind team is considering with their launch,<00:02:50.160> is<00:02:50.400> Muse<00:02:50.720> Spark<00:02:51.120> 1.2,<00:02:52.080> which<00:02:52.319> is 00:02:52.470 --> 00:02:52.480 align:start position:0% launch, is Muse Spark 1.2, which is 00:02:52.480 --> 00:02:54.630 align:start position:0% launch, is Muse Spark 1.2, which is sitting<00:02:52.720> at<00:02:52.959> 54.9%. 00:02:54.630 --> 00:02:54.640 align:start position:0% sitting at 54.9%. 00:02:54.640 --> 00:02:57.509 align:start position:0% sitting at 54.9%. So,<00:02:54.879> welcome<00:02:55.280> Meta<00:02:55.680> to<00:02:55.840> the<00:02:56.000> benchmark<00:02:56.959> charts 00:02:57.509 --> 00:02:57.519 align:start position:0% So, welcome Meta to the benchmark charts 00:02:57.519 --> 00:02:59.350 align:start position:0% So, welcome Meta to the benchmark charts because<00:02:57.920> now<00:02:58.319> we're<00:02:58.560> starting<00:02:58.800> to<00:02:58.959> see<00:02:59.120> it 00:02:59.350 --> 00:02:59.360 align:start position:0% because now we're starting to see it 00:02:59.360 --> 00:03:01.509 align:start position:0% because now we're starting to see it appear<00:02:59.599> on<00:02:59.840> more<00:03:00.000> and<00:03:00.160> more<00:03:00.319> benchmarks.<00:03:01.360> And 00:03:01.509 --> 00:03:01.519 align:start position:0% appear on more and more benchmarks. And 00:03:01.519 --> 00:03:02.790 align:start position:0% appear on more and more benchmarks. And if<00:03:01.599> we<00:03:01.760> also<00:03:02.000> take<00:03:02.080> a<00:03:02.239> look<00:03:02.319> at<00:03:02.560> web 00:03:02.790 --> 00:03:02.800 align:start position:0% if we also take a look at web 00:03:02.800 --> 00:03:05.350 align:start position:0% if we also take a look at web development,<00:03:03.680> this<00:03:03.840> model<00:03:04.239> is<00:03:04.560> getting<00:03:04.800> a<00:03:04.959> ELO 00:03:05.350 --> 00:03:05.360 align:start position:0% development, this model is getting a ELO 00:03:05.360 --> 00:03:08.149 align:start position:0% development, this model is getting a ELO score<00:03:05.599> of<00:03:05.920> 1588<00:03:06.720> versus<00:03:07.120> their<00:03:07.360> old<00:03:07.599> model<00:03:07.920> is 00:03:08.149 --> 00:03:08.159 align:start position:0% score of 1588 versus their old model is 00:03:08.159 --> 00:03:10.309 align:start position:0% score of 1588 versus their old model is 1538.<00:03:08.879> So,<00:03:09.040> not<00:03:09.200> a<00:03:09.360> crazy<00:03:09.680> difference,<00:03:10.000> but 00:03:10.309 --> 00:03:10.319 align:start position:0% 1538. So, not a crazy difference, but 00:03:10.319 --> 00:03:11.830 align:start position:0% 1538. So, not a crazy difference, but compared<00:03:10.640> to<00:03:10.800> everything<00:03:11.120> else<00:03:11.360> out<00:03:11.519> there, 00:03:11.830 --> 00:03:11.840 align:start position:0% compared to everything else out there, 00:03:11.840 --> 00:03:13.350 align:start position:0% compared to everything else out there, this<00:03:12.000> is<00:03:12.159> number<00:03:12.400> one.<00:03:12.879> When<00:03:13.040> I<00:03:13.200> say 00:03:13.350 --> 00:03:13.360 align:start position:0% this is number one. When I say 00:03:13.360 --> 00:03:14.869 align:start position:0% this is number one. When I say everything<00:03:13.599> else<00:03:13.840> out<00:03:14.080> there,<00:03:14.319> once<00:03:14.560> again, 00:03:14.869 --> 00:03:14.879 align:start position:0% everything else out there, once again, 00:03:14.879 --> 00:03:16.710 align:start position:0% everything else out there, once again, compared<00:03:15.200> to<00:03:15.360> all<00:03:15.519> the<00:03:15.680> mid-tier<00:03:16.239> models, 00:03:16.710 --> 00:03:16.720 align:start position:0% compared to all the mid-tier models, 00:03:16.720 --> 00:03:18.630 align:start position:0% compared to all the mid-tier models, this<00:03:16.879> is<00:03:17.200> beating<00:03:17.760> all<00:03:17.920> of<00:03:18.080> them<00:03:18.319> when<00:03:18.560> it 00:03:18.630 --> 00:03:18.640 align:start position:0% this is beating all of them when it 00:03:18.640 --> 00:03:20.630 align:start position:0% this is beating all of them when it comes<00:03:18.800> to<00:03:18.959> web<00:03:19.280> development.<00:03:20.159> Now,<00:03:20.400> this 00:03:20.630 --> 00:03:20.640 align:start position:0% comes to web development. Now, this 00:03:20.640 --> 00:03:22.470 align:start position:0% comes to web development. Now, this model<00:03:20.959> is<00:03:21.120> probably<00:03:21.360> going<00:03:21.519> to<00:03:21.599> be<00:03:21.840> used<00:03:22.080> in 00:03:22.470 --> 00:03:22.480 align:start position:0% model is probably going to be used in 00:03:22.480 --> 00:03:24.149 align:start position:0% model is probably going to be used in enterprises<00:03:23.200> a<00:03:23.360> lot<00:03:23.519> just<00:03:23.760> because<00:03:23.920> of 00:03:24.149 --> 00:03:24.159 align:start position:0% enterprises a lot just because of 00:03:24.159 --> 00:03:25.990 align:start position:0% enterprises a lot just because of Google's<00:03:24.640> footprint<00:03:25.040> in<00:03:25.280> the<00:03:25.440> enterprise 00:03:25.990 --> 00:03:26.000 align:start position:0% Google's footprint in the enterprise 00:03:26.000 --> 00:03:28.149 align:start position:0% Google's footprint in the enterprise space,<00:03:26.480> but<00:03:26.720> we<00:03:26.959> are<00:03:27.120> seeing<00:03:27.280> this<00:03:27.519> model 00:03:28.149 --> 00:03:28.159 align:start position:0% space, but we are seeing this model 00:03:28.159 --> 00:03:31.589 align:start position:0% space, but we are seeing this model achieve<00:03:28.720> on<00:03:28.959> the<00:03:29.120> automation<00:03:29.680> bench<00:03:30.080> 30.4%. 00:03:31.589 --> 00:03:31.599 align:start position:0% achieve on the automation bench 30.4%. 00:03:31.599 --> 00:03:35.430 align:start position:0% achieve on the automation bench 30.4%. And<00:03:31.840> GPT<00:03:32.319> 5.6<00:03:32.959> Terra<00:03:33.360> sits<00:03:33.599> at<00:03:33.840> 23.6%. 00:03:35.430 --> 00:03:35.440 align:start position:0% And GPT 5.6 Terra sits at 23.6%. 00:03:35.440 --> 00:03:37.589 align:start position:0% And GPT 5.6 Terra sits at 23.6%. So,<00:03:35.680> yes,<00:03:36.000> this<00:03:36.159> model<00:03:36.400> is<00:03:36.640> stronger<00:03:37.040> than<00:03:37.360> the 00:03:37.589 --> 00:03:37.599 align:start position:0% So, yes, this model is stronger than the 00:03:37.599 --> 00:03:39.750 align:start position:0% So, yes, this model is stronger than the other<00:03:37.920> Flash<00:03:38.319> models<00:03:38.640> out<00:03:38.799> there,<00:03:39.200> but<00:03:39.440> as<00:03:39.599> I 00:03:39.750 --> 00:03:39.760 align:start position:0% other Flash models out there, but as I 00:03:39.760 --> 00:03:41.430 align:start position:0% other Flash models out there, but as I said,<00:03:39.920> they<00:03:40.159> haven't<00:03:40.400> included<00:03:40.879> Deepseek 00:03:41.430 --> 00:03:41.440 align:start position:0% said, they haven't included Deepseek 00:03:41.440 --> 00:03:43.589 align:start position:0% said, they haven't included Deepseek version<00:03:41.680> for<00:03:41.920> Flash<00:03:42.319> because<00:03:42.560> if<00:03:42.720> they<00:03:42.959> do,<00:03:43.360> in 00:03:43.589 --> 00:03:43.599 align:start position:0% version for Flash because if they do, in 00:03:43.599 --> 00:03:45.509 align:start position:0% version for Flash because if they do, in some<00:03:43.840> areas,<00:03:44.239> that<00:03:44.480> model<00:03:44.720> is<00:03:44.959> actually<00:03:45.200> quite 00:03:45.509 --> 00:03:45.519 align:start position:0% some areas, that model is actually quite 00:03:45.519 --> 00:03:48.470 align:start position:0% some areas, that model is actually quite better<00:03:45.680> than<00:03:45.840> the<00:03:46.080> 3.7<00:03:46.799> Flash<00:03:47.120> model.<00:03:48.159> Before 00:03:48.470 --> 00:03:48.480 align:start position:0% better than the 3.7 Flash model. Before 00:03:48.480 --> 00:03:50.309 align:start position:0% better than the 3.7 Flash model. Before we<00:03:48.720> continue,<00:03:49.360> if<00:03:49.599> you're<00:03:49.760> building<00:03:50.000> AI 00:03:50.309 --> 00:03:50.319 align:start position:0% we continue, if you're building AI 00:03:50.319 --> 00:03:52.470 align:start position:0% we continue, if you're building AI agents<00:03:50.720> or<00:03:51.040> just<00:03:51.200> messing<00:03:51.519> around<00:03:51.680> with<00:03:51.920> them, 00:03:52.470 --> 00:03:52.480 align:start position:0% agents or just messing around with them, 00:03:52.480 --> 00:03:54.789 align:start position:0% agents or just messing around with them, Arcade<00:03:53.040> is<00:03:53.200> worth<00:03:53.519> knowing<00:03:53.760> about.<00:03:54.400> It's<00:03:54.560> the 00:03:54.789 --> 00:03:54.799 align:start position:0% Arcade is worth knowing about. It's the 00:03:54.799 --> 00:03:56.789 align:start position:0% Arcade is worth knowing about. It's the runtime<00:03:55.120> that<00:03:55.280> lets<00:03:55.519> your<00:03:55.680> agent<00:03:56.159> actually<00:03:56.560> do 00:03:56.789 --> 00:03:56.799 align:start position:0% runtime that lets your agent actually do 00:03:56.799 --> 00:03:58.309 align:start position:0% runtime that lets your agent actually do things<00:03:57.120> instead<00:03:57.439> of<00:03:57.599> just<00:03:57.840> talking<00:03:58.080> about 00:03:58.309 --> 00:03:58.319 align:start position:0% things instead of just talking about 00:03:58.319 --> 00:04:00.550 align:start position:0% things instead of just talking about them<00:03:59.040> because<00:03:59.280> that's<00:03:59.519> the<00:03:59.680> gap<00:04:00.000> right<00:04:00.159> now. 00:04:00.550 --> 00:04:00.560 align:start position:0% them because that's the gap right now. 00:04:00.560 --> 00:04:02.789 align:start position:0% them because that's the gap right now. The<00:04:00.799> models<00:04:01.200> are<00:04:01.439> smart<00:04:01.680> enough.<00:04:02.239> Your<00:04:02.480> agent 00:04:02.789 --> 00:04:02.799 align:start position:0% The models are smart enough. Your agent 00:04:02.799 --> 00:04:04.710 align:start position:0% The models are smart enough. Your agent can<00:04:03.040> figure<00:04:03.200> out<00:04:03.439> exactly<00:04:04.000> what<00:04:04.319> needs<00:04:04.560> to 00:04:04.710 --> 00:04:04.720 align:start position:0% can figure out exactly what needs to 00:04:04.720 --> 00:04:06.550 align:start position:0% can figure out exactly what needs to happen<00:04:04.879> in<00:04:05.120> your<00:04:05.360> email,<00:04:05.760> your<00:04:05.920> Slack,<00:04:06.400> your 00:04:06.550 --> 00:04:06.560 align:start position:0% happen in your email, your Slack, your 00:04:06.560 --> 00:04:09.670 align:start position:0% happen in your email, your Slack, your CRM.<00:04:07.439> It<00:04:07.680> just<00:04:07.920> can't<00:04:08.239> go<00:04:08.400> in<00:04:08.640> and<00:04:08.879> do<00:04:09.040> it.<00:04:09.519> And 00:04:09.670 --> 00:04:09.680 align:start position:0% CRM. It just can't go in and do it. And 00:04:09.680 --> 00:04:11.509 align:start position:0% CRM. It just can't go in and do it. And the<00:04:09.840> reason<00:04:10.080> isn't<00:04:10.480> intelligence,<00:04:11.200> it's 00:04:11.509 --> 00:04:11.519 align:start position:0% the reason isn't intelligence, it's 00:04:11.519 --> 00:04:13.670 align:start position:0% the reason isn't intelligence, it's permissions.<00:04:12.560> Something<00:04:12.879> has<00:04:13.040> to<00:04:13.200> prove<00:04:13.439> the 00:04:13.670 --> 00:04:13.680 align:start position:0% permissions. Something has to prove the 00:04:13.680 --> 00:04:15.830 align:start position:0% permissions. Something has to prove the agent<00:04:14.000> is<00:04:14.239> allowed<00:04:14.480> to<00:04:14.640> act<00:04:14.879> on<00:04:15.040> behalf<00:04:15.360> of<00:04:15.519> a 00:04:15.830 --> 00:04:15.840 align:start position:0% agent is allowed to act on behalf of a 00:04:15.840 --> 00:04:18.469 align:start position:0% agent is allowed to act on behalf of a specific<00:04:16.239> person<00:04:16.720> in<00:04:17.040> a<00:04:17.199> specific<00:04:17.680> account. 00:04:18.469 --> 00:04:18.479 align:start position:0% specific person in a specific account. 00:04:18.479 --> 00:04:20.229 align:start position:0% specific person in a specific account. That's<00:04:18.799> the<00:04:18.959> messy<00:04:19.359> part<00:04:19.600> everyone<00:04:20.000> runs 00:04:20.229 --> 00:04:20.239 align:start position:0% That's the messy part everyone runs 00:04:20.239 --> 00:04:22.469 align:start position:0% That's the messy part everyone runs into,<00:04:20.720> and<00:04:20.959> it's<00:04:21.199> the<00:04:21.359> partit<00:04:22.160> actually 00:04:22.469 --> 00:04:22.479 align:start position:0% into, and it's the partit actually 00:04:22.479 --> 00:04:24.310 align:start position:0% into, and it's the partit actually handles<00:04:22.880> for<00:04:23.040> you.<00:04:23.520> So<00:04:23.759> instead<00:04:24.000> of<00:04:24.160> your 00:04:24.310 --> 00:04:24.320 align:start position:0% handles for you. So instead of your 00:04:24.320 --> 00:04:26.230 align:start position:0% handles for you. So instead of your agent<00:04:24.639> using<00:04:24.960> one<00:04:25.280> shared<00:04:25.600> login<00:04:26.000> for 00:04:26.230 --> 00:04:26.240 align:start position:0% agent using one shared login for 00:04:26.240 --> 00:04:28.150 align:start position:0% agent using one shared login for everybody,<00:04:26.960> it<00:04:27.120> acts<00:04:27.360> as<00:04:27.600> whoever<00:04:27.919> is 00:04:28.150 --> 00:04:28.160 align:start position:0% everybody, it acts as whoever is 00:04:28.160 --> 00:04:30.310 align:start position:0% everybody, it acts as whoever is actually<00:04:28.560> signed<00:04:28.880> in<00:04:29.280> with<00:04:29.600> exactly<00:04:30.000> the 00:04:30.310 --> 00:04:30.320 align:start position:0% actually signed in with exactly the 00:04:30.320 --> 00:04:32.550 align:start position:0% actually signed in with exactly the access<00:04:30.639> that<00:04:30.960> person<00:04:31.280> has.<00:04:32.000> If<00:04:32.160> they<00:04:32.400> can't 00:04:32.550 --> 00:04:32.560 align:start position:0% access that person has. If they can't 00:04:32.560 --> 00:04:34.870 align:start position:0% access that person has. If they can't see<00:04:32.800> something,<00:04:33.440> the<00:04:33.680> agent<00:04:34.000> can't<00:04:34.320> either. 00:04:34.870 --> 00:04:34.880 align:start position:0% see something, the agent can't either. 00:04:34.880 --> 00:04:36.469 align:start position:0% see something, the agent can't either. And<00:04:35.040> you<00:04:35.199> never<00:04:35.440> have<00:04:35.600> to<00:04:35.759> touch<00:04:36.000> any<00:04:36.160> of<00:04:36.320> that 00:04:36.469 --> 00:04:36.479 align:start position:0% And you never have to touch any of that 00:04:36.479 --> 00:04:38.950 align:start position:0% And you never have to touch any of that setup<00:04:36.880> yourself.<00:04:37.600> Then<00:04:37.919> there's<00:04:38.080> the<00:04:38.320> tools. 00:04:38.950 --> 00:04:38.960 align:start position:0% setup yourself. Then there's the tools. 00:04:38.960 --> 00:04:40.790 align:start position:0% setup yourself. Then there's the tools. Arcade<00:04:39.440> has<00:04:39.680> thousands<00:04:40.080> of<00:04:40.240> them<00:04:40.479> already 00:04:40.790 --> 00:04:40.800 align:start position:0% Arcade has thousands of them already 00:04:40.800 --> 00:04:43.110 align:start position:0% Arcade has thousands of them already built<00:04:41.040> for<00:04:41.280> Gmail,<00:04:41.840> Google<00:04:42.080> Drive,<00:04:42.560> Slack, 00:04:43.110 --> 00:04:43.120 align:start position:0% built for Gmail, Google Drive, Slack, 00:04:43.120 --> 00:04:45.510 align:start position:0% built for Gmail, Google Drive, Slack, Notion,<00:04:43.680> Salesforce,<00:04:44.720> most<00:04:44.880> of<00:04:45.040> the<00:04:45.199> apps 00:04:45.510 --> 00:04:45.520 align:start position:0% Notion, Salesforce, most of the apps 00:04:45.520 --> 00:04:47.350 align:start position:0% Notion, Salesforce, most of the apps people<00:04:45.840> already<00:04:46.160> work<00:04:46.400> in,<00:04:46.800> and<00:04:47.040> they<00:04:47.280> are 00:04:47.350 --> 00:04:47.360 align:start position:0% people already work in, and they are 00:04:47.360 --> 00:04:49.749 align:start position:0% people already work in, and they are built<00:04:47.680> specifically<00:04:48.160> for<00:04:48.400> AI<00:04:48.800> to<00:04:48.960> use.<00:04:49.440> So<00:04:49.600> the 00:04:49.749 --> 00:04:49.759 align:start position:0% built specifically for AI to use. So the 00:04:49.759 --> 00:04:51.430 align:start position:0% built specifically for AI to use. So the agent<00:04:50.080> gets<00:04:50.320> it<00:04:50.560> right<00:04:50.720> the<00:04:50.960> first<00:04:51.199> time 00:04:51.430 --> 00:04:51.440 align:start position:0% agent gets it right the first time 00:04:51.440 --> 00:04:53.030 align:start position:0% agent gets it right the first time instead<00:04:51.759> of<00:04:51.919> guessing<00:04:52.240> and<00:04:52.479> failing<00:04:52.800> and 00:04:53.030 --> 00:04:53.040 align:start position:0% instead of guessing and failing and 00:04:53.040 --> 00:04:55.110 align:start position:0% instead of guessing and failing and retrying.<00:04:53.840> It<00:04:54.080> also<00:04:54.320> keeps<00:04:54.560> a<00:04:54.720> record<00:04:54.960> of 00:04:55.110 --> 00:04:55.120 align:start position:0% retrying. It also keeps a record of 00:04:55.120 --> 00:04:57.510 align:start position:0% retrying. It also keeps a record of everything,<00:04:55.840> what<00:04:56.080> the<00:04:56.320> agent<00:04:56.639> did,<00:04:56.960> for<00:04:57.199> who 00:04:57.510 --> 00:04:57.520 align:start position:0% everything, what the agent did, for who 00:04:57.520 --> 00:04:59.430 align:start position:0% everything, what the agent did, for who and<00:04:57.759> where,<00:04:58.240> which<00:04:58.560> matters<00:04:58.880> a<00:04:59.040> lot<00:04:59.199> the 00:04:59.430 --> 00:04:59.440 align:start position:0% and where, which matters a lot the 00:04:59.440 --> 00:05:00.870 align:start position:0% and where, which matters a lot the moment<00:04:59.680> other<00:05:00.000> people<00:05:00.240> start<00:05:00.479> using<00:05:00.720> the 00:05:00.870 --> 00:05:00.880 align:start position:0% moment other people start using the 00:05:00.880 --> 00:05:02.629 align:start position:0% moment other people start using the thing<00:05:01.120> you<00:05:01.280> built.<00:05:01.840> And<00:05:02.000> it<00:05:02.240> works<00:05:02.400> with 00:05:02.629 --> 00:05:02.639 align:start position:0% thing you built. And it works with 00:05:02.639 --> 00:05:04.550 align:start position:0% thing you built. And it works with whatever<00:05:03.040> you're<00:05:03.280> already<00:05:03.600> using.<00:05:04.240> Any 00:05:04.550 --> 00:05:04.560 align:start position:0% whatever you're already using. Any 00:05:04.560 --> 00:05:06.950 align:start position:0% whatever you're already using. Any model,<00:05:04.960> any<00:05:05.280> framework,<00:05:05.919> cloud,<00:05:06.400> cursor, 00:05:06.950 --> 00:05:06.960 align:start position:0% model, any framework, cloud, cursor, 00:05:06.960 --> 00:05:09.510 align:start position:0% model, any framework, cloud, cursor, chat,<00:05:07.280> GPT,<00:05:08.080> doesn't<00:05:08.400> matter.<00:05:09.039> So<00:05:09.199> you're<00:05:09.360> not 00:05:09.510 --> 00:05:09.520 align:start position:0% chat, GPT, doesn't matter. So you're not 00:05:09.520 --> 00:05:11.129 align:start position:0% chat, GPT, doesn't matter. So you're not just<00:05:09.680> giving<00:05:09.840> an<00:05:10.080> AI<00:05:10.400> a<00:05:10.560> list<00:05:10.720> of<00:05:10.880> tools<00:05:11.120> and 00:05:11.129 --> 00:05:11.139 align:start position:0% just giving an AI a list of tools and 00:05:11.139 --> 00:05:12.710 align:start position:0% just giving an AI a list of tools and [music]<00:05:11.360> hoping<00:05:11.600> it<00:05:11.840> works.<00:05:12.320> You're<00:05:12.560> giving 00:05:12.710 --> 00:05:12.720 align:start position:0% [music] hoping it works. You're giving 00:05:12.720 --> 00:05:14.790 align:start position:0% [music] hoping it works. You're giving it<00:05:12.880> a<00:05:13.120> place<00:05:13.199> where<00:05:13.440> it<00:05:13.600> can<00:05:13.759> safely<00:05:14.240> take<00:05:14.479> real 00:05:14.790 --> 00:05:14.800 align:start position:0% it a place where it can safely take real 00:05:14.800 --> 00:05:17.350 align:start position:0% it a place where it can safely take real actions<00:05:15.280> in<00:05:15.600> real<00:05:15.840> apps.<00:05:16.560> It's<00:05:16.800> free<00:05:16.960> to<00:05:17.120> start 00:05:17.350 --> 00:05:17.360 align:start position:0% actions in real apps. It's free to start 00:05:17.360 --> 00:05:19.110 align:start position:0% actions in real apps. It's free to start and<00:05:17.520> the<00:05:17.759> link<00:05:17.919> is<00:05:18.080> in<00:05:18.240> the<00:05:18.400> description. 00:05:19.110 --> 00:05:19.120 align:start position:0% and the link is in the description. 00:05:19.120 --> 00:05:20.870 align:start position:0% and the link is in the description. Thank<00:05:19.280> you<00:05:19.520> once<00:05:19.759> again<00:05:19.919> for<00:05:20.160> Arcade<00:05:20.639> for 00:05:20.870 --> 00:05:20.880 align:start position:0% Thank you once again for Arcade for 00:05:20.880 --> 00:05:22.950 align:start position:0% Thank you once again for Arcade for sponsoring<00:05:21.280> today's<00:05:21.680> video.<00:05:22.240> Now,<00:05:22.560> let's<00:05:22.800> get 00:05:22.950 --> 00:05:22.960 align:start position:0% sponsoring today's video. Now, let's get 00:05:22.960 --> 00:05:25.430 align:start position:0% sponsoring today's video. Now, let's get back<00:05:23.120> into<00:05:23.360> the<00:05:23.600> video.<00:05:24.639> Now,<00:05:24.960> one<00:05:25.120> thing<00:05:25.280> to 00:05:25.430 --> 00:05:25.440 align:start position:0% back into the video. Now, one thing to 00:05:25.440 --> 00:05:28.550 align:start position:0% back into the video. Now, one thing to note<00:05:25.680> is<00:05:25.919> that<00:05:26.400> the<00:05:26.639> Gemini<00:05:27.120> 3.7<00:05:27.759> Flash<00:05:28.080> model 00:05:28.550 --> 00:05:28.560 align:start position:0% note is that the Gemini 3.7 Flash model 00:05:28.560 --> 00:05:30.310 align:start position:0% note is that the Gemini 3.7 Flash model through<00:05:28.800> the<00:05:29.039> end<00:05:29.120> of<00:05:29.280> this<00:05:29.440> year,<00:05:29.680> so<00:05:29.919> end<00:05:30.080> of 00:05:30.310 --> 00:05:30.320 align:start position:0% through the end of this year, so end of 00:05:30.320 --> 00:05:33.350 align:start position:0% through the end of this year, so end of 2026,<00:05:31.520> they<00:05:31.840> have<00:05:32.000> a<00:05:32.320> cheap<00:05:32.720> pricing<00:05:33.120> model 00:05:33.350 --> 00:05:33.360 align:start position:0% 2026, they have a cheap pricing model 00:05:33.360 --> 00:05:36.150 align:start position:0% 2026, they have a cheap pricing model that<00:05:33.600> they're<00:05:33.840> placing<00:05:34.160> on<00:05:34.320> the<00:05:34.560> model.<00:05:35.280> 75 00:05:36.150 --> 00:05:36.160 align:start position:0% that they're placing on the model. 75 00:05:36.160 --> 00:05:39.830 align:start position:0% that they're placing on the model. 75 per<00:05:36.639> 1<00:05:36.800> million<00:05:37.120> input<00:05:37.440> tokens<00:05:37.840> and<00:05:38.240> $3.75<00:05:39.680> for 00:05:39.830 --> 00:05:39.840 align:start position:0% per 1 million input tokens and $3.75 for 00:05:39.840 --> 00:05:41.670 align:start position:0% per 1 million input tokens and $3.75 for 1<00:05:40.080> million<00:05:40.320> output<00:05:40.639> tokens.<00:05:41.120> So,<00:05:41.280> it's<00:05:41.440> a 00:05:41.670 --> 00:05:41.680 align:start position:0% 1 million output tokens. So, it's a 00:05:41.680 --> 00:05:43.749 align:start position:0% 1 million output tokens. So, it's a competitive<00:05:42.160> price,<00:05:42.479> but<00:05:42.720> this<00:05:42.960> is<00:05:43.120> only<00:05:43.440> for 00:05:43.749 --> 00:05:43.759 align:start position:0% competitive price, but this is only for 00:05:43.759 --> 00:05:45.909 align:start position:0% competitive price, but this is only for the<00:05:43.919> next<00:05:44.160> 6<00:05:44.400> months.<00:05:45.039> Because<00:05:45.360> after<00:05:45.680> those 00:05:45.909 --> 00:05:45.919 align:start position:0% the next 6 months. Because after those 00:05:45.919 --> 00:05:48.230 align:start position:0% the next 6 months. Because after those six<00:05:46.240> months<00:05:46.560> are<00:05:46.720> done,<00:05:47.199> the<00:05:47.440> model's<00:05:47.840> pricing 00:05:48.230 --> 00:05:48.240 align:start position:0% six months are done, the model's pricing 00:05:48.240 --> 00:05:50.150 align:start position:0% six months are done, the model's pricing is<00:05:48.479> actually,<00:05:49.199> you<00:05:49.360> know,<00:05:49.440> a<00:05:49.680> little<00:05:49.840> bit<00:05:49.919> more 00:05:50.150 --> 00:05:50.160 align:start position:0% is actually, you know, a little bit more 00:05:50.160 --> 00:05:51.909 align:start position:0% is actually, you know, a little bit more expensive.<00:05:50.720> And<00:05:50.880> now<00:05:51.120> they<00:05:51.280> show<00:05:51.440> it<00:05:51.600> at<00:05:51.759> the 00:05:51.909 --> 00:05:51.919 align:start position:0% expensive. And now they show it at the 00:05:51.919 --> 00:05:54.710 align:start position:0% expensive. And now they show it at the bottom<00:05:52.160> over<00:05:52.479> here,<00:05:53.120> you<00:05:53.280> can<00:05:53.440> see<00:05:53.600> that<00:05:54.240> after 00:05:54.710 --> 00:05:54.720 align:start position:0% bottom over here, you can see that after 00:05:54.720 --> 00:05:58.150 align:start position:0% bottom over here, you can see that after starting<00:05:55.120> January<00:05:55.600> of<00:05:55.840> 2027,<00:05:57.199> it<00:05:57.440> will<00:05:57.600> become 00:05:58.150 --> 00:05:58.160 align:start position:0% starting January of 2027, it will become 00:05:58.160 --> 00:06:02.469 align:start position:0% starting January of 2027, it will become $1.50<00:05:59.199> per<00:05:59.440> input<00:05:59.759> and<00:06:00.160> $7.50<00:06:01.120> per<00:06:01.360> output.<00:06:02.000> So 00:06:02.469 --> 00:06:02.479 align:start position:0% $1.50 per input and $7.50 per output. So 00:06:02.479 --> 00:06:04.710 align:start position:0% $1.50 per input and $7.50 per output. So yeah,<00:06:02.720> it's<00:06:03.199> still<00:06:03.520> cheap<00:06:03.919> compared<00:06:04.319> to<00:06:04.560> the 00:06:04.710 --> 00:06:04.720 align:start position:0% yeah, it's still cheap compared to the 00:06:04.720 --> 00:06:06.550 align:start position:0% yeah, it's still cheap compared to the other<00:06:04.960> frontier<00:06:05.440> labs,<00:06:05.759> but<00:06:05.919> it's<00:06:06.160> not<00:06:06.319> as 00:06:06.550 --> 00:06:06.560 align:start position:0% other frontier labs, but it's not as 00:06:06.560 --> 00:06:08.629 align:start position:0% other frontier labs, but it's not as cheap<00:06:06.800> as,<00:06:07.120> for<00:06:07.120> example,<00:06:07.520> MU<00:06:07.919> Spark<00:06:08.240> when<00:06:08.479> the 00:06:08.629 --> 00:06:08.639 align:start position:0% cheap as, for example, MU Spark when the 00:06:08.639 --> 00:06:11.189 align:start position:0% cheap as, for example, MU Spark when the pricing<00:06:08.960> is<00:06:09.199> updated<00:06:10.000> or<00:06:10.240> even<00:06:10.479> the<00:06:10.639> Deepsee 00:06:11.189 --> 00:06:11.199 align:start position:0% pricing is updated or even the Deepsee 00:06:11.199 --> 00:06:13.270 align:start position:0% pricing is updated or even the Deepsee version<00:06:11.520> for<00:06:11.759> Flash.<00:06:12.479> But<00:06:12.720> across<00:06:13.039> these 00:06:13.270 --> 00:06:13.280 align:start position:0% version for Flash. But across these 00:06:13.280 --> 00:06:15.430 align:start position:0% version for Flash. But across these benchmarks,<00:06:13.840> we<00:06:14.080> can<00:06:14.160> see<00:06:14.319> that<00:06:14.560> this<00:06:14.880> model 00:06:15.430 --> 00:06:15.440 align:start position:0% benchmarks, we can see that this model 00:06:15.440 --> 00:06:17.350 align:start position:0% benchmarks, we can see that this model is<00:06:15.759> better<00:06:16.000> in<00:06:16.240> many<00:06:16.479> areas<00:06:16.880> that<00:06:17.120> they 00:06:17.350 --> 00:06:17.360 align:start position:0% is better in many areas that they 00:06:17.360 --> 00:06:18.870 align:start position:0% is better in many areas that they highlighted<00:06:17.759> at<00:06:17.919> the<00:06:18.080> top.<00:06:18.400> But<00:06:18.479> then<00:06:18.639> they 00:06:18.870 --> 00:06:18.880 align:start position:0% highlighted at the top. But then they 00:06:18.880 --> 00:06:21.189 align:start position:0% highlighted at the top. But then they also<00:06:19.120> have<00:06:19.280> some<00:06:19.520> other<00:06:19.759> areas<00:06:20.160> like<00:06:20.880> long 00:06:21.189 --> 00:06:21.199 align:start position:0% also have some other areas like long 00:06:21.199 --> 00:06:22.950 align:start position:0% also have some other areas like long video<00:06:21.600> understanding,<00:06:22.160> which<00:06:22.400> this<00:06:22.639> model 00:06:22.950 --> 00:06:22.960 align:start position:0% video understanding, which this model 00:06:22.960 --> 00:06:27.189 align:start position:0% video understanding, which this model excels<00:06:23.360> at<00:06:23.600> 85.4%<00:06:25.039> versus<00:06:25.440> 78.9% 00:06:27.189 --> 00:06:27.199 align:start position:0% excels at 85.4% versus 78.9% 00:06:27.199 --> 00:06:29.110 align:start position:0% excels at 85.4% versus 78.9% for<00:06:27.520> the<00:06:27.680> Terra<00:06:28.160> model.<00:06:28.400> And<00:06:28.560> then<00:06:28.720> the<00:06:28.880> old 00:06:29.110 --> 00:06:29.120 align:start position:0% for the Terra model. And then the old 00:06:29.120 --> 00:06:30.870 align:start position:0% for the Terra model. And then the old model<00:06:29.440> was<00:06:29.680> also<00:06:30.000> pretty<00:06:30.319> good<00:06:30.400> at<00:06:30.639> that, 00:06:30.870 --> 00:06:30.880 align:start position:0% model was also pretty good at that, 00:06:30.880 --> 00:06:32.550 align:start position:0% model was also pretty good at that, 84.2%. 00:06:32.550 --> 00:06:32.560 align:start position:0% 84.2%. 00:06:32.560 --> 00:06:34.790 align:start position:0% 84.2%. Then<00:06:32.880> long<00:06:33.199> context<00:06:33.759> performance<00:06:34.400> the<00:06:34.560> model 00:06:34.790 --> 00:06:34.800 align:start position:0% Then long context performance the model 00:06:34.800 --> 00:06:39.430 align:start position:0% Then long context performance the model is<00:06:34.960> at<00:06:35.280> 97%<00:06:36.560> and<00:06:36.800> this<00:06:37.039> model<00:06:37.520> the<00:06:37.840> GPT<00:06:38.800> 6<00:06:39.039> Terra 00:06:39.430 --> 00:06:39.440 align:start position:0% is at 97% and this model the GPT 6 Terra 00:06:39.440 --> 00:06:42.150 align:start position:0% is at 97% and this model the GPT 6 Terra one<00:06:39.919> it<00:06:40.080> sits<00:06:40.400> at<00:06:40.560> 93.5%. 00:06:42.150 --> 00:06:42.160 align:start position:0% one it sits at 93.5%. 00:06:42.160 --> 00:06:44.150 align:start position:0% one it sits at 93.5%. So<00:06:42.400> yeah<00:06:42.639> this<00:06:42.800> model<00:06:43.039> in<00:06:43.199> summary<00:06:43.759> it<00:06:44.000> is 00:06:44.150 --> 00:06:44.160 align:start position:0% So yeah this model in summary it is 00:06:44.160 --> 00:06:46.469 align:start position:0% So yeah this model in summary it is better<00:06:44.479> so<00:06:44.720> it's<00:06:44.960> not<00:06:45.199> all<00:06:45.440> negative<00:06:45.919> but<00:06:46.240> it's 00:06:46.469 --> 00:06:46.479 align:start position:0% better so it's not all negative but it's 00:06:46.479 --> 00:06:48.629 align:start position:0% better so it's not all negative but it's not<00:06:46.639> all<00:06:46.880> like<00:06:47.199> you<00:06:47.360> know<00:06:47.840> that<00:06:48.160> positive 00:06:48.629 --> 00:06:48.639 align:start position:0% not all like you know that positive 00:06:48.639 --> 00:06:50.469 align:start position:0% not all like you know that positive where<00:06:48.960> you<00:06:49.199> are<00:06:49.360> super<00:06:49.680> excited<00:06:50.000> for<00:06:50.160> Google 00:06:50.469 --> 00:06:50.479 align:start position:0% where you are super excited for Google 00:06:50.479 --> 00:06:52.230 align:start position:0% where you are super excited for Google Deep<00:06:50.800> Mind<00:06:51.039> because<00:06:51.280> as<00:06:51.440> I<00:06:51.600> said<00:06:52.000> they're 00:06:52.230 --> 00:06:52.240 align:start position:0% Deep Mind because as I said they're 00:06:52.240 --> 00:06:53.830 align:start position:0% Deep Mind because as I said they're probably<00:06:52.560> still<00:06:52.800> holding<00:06:53.120> off<00:06:53.360> their<00:06:53.520> biggest 00:06:53.830 --> 00:06:53.840 align:start position:0% probably still holding off their biggest 00:06:53.840 --> 00:06:56.309 align:start position:0% probably still holding off their biggest release<00:06:54.080> for<00:06:54.240> Gemini<00:06:54.720> 4<00:06:54.960> lineup.<00:06:55.840> Now,<00:06:56.160> one 00:06:56.309 --> 00:06:56.319 align:start position:0% release for Gemini 4 lineup. Now, one 00:06:56.319 --> 00:06:58.150 align:start position:0% release for Gemini 4 lineup. Now, one thing<00:06:56.479> people<00:06:57.120> might<00:06:57.360> have<00:06:57.520> missed<00:06:57.759> in<00:06:57.919> their 00:06:58.150 --> 00:06:58.160 align:start position:0% thing people might have missed in their 00:06:58.160 --> 00:06:59.749 align:start position:0% thing people might have missed in their charts<00:06:58.479> because<00:06:58.720> these<00:06:59.039> charts<00:06:59.360> sometimes 00:06:59.749 --> 00:06:59.759 align:start position:0% charts because these charts sometimes 00:06:59.759 --> 00:07:02.790 align:start position:0% charts because these charts sometimes are<00:07:00.000> so<00:07:00.479> messy<00:07:00.960> to<00:07:01.280> read<00:07:01.520> and<00:07:01.840> understand,<00:07:02.319> but 00:07:02.790 --> 00:07:02.800 align:start position:0% are so messy to read and understand, but 00:07:02.800 --> 00:07:06.150 align:start position:0% are so messy to read and understand, but the<00:07:03.039> 3.7<00:07:03.759> flash<00:07:04.160> is<00:07:04.400> worse<00:07:04.720> than<00:07:04.880> GPT<00:07:05.520> 5.6 00:07:06.150 --> 00:07:06.160 align:start position:0% the 3.7 flash is worse than GPT 5.6 00:07:06.160 --> 00:07:08.150 align:start position:0% the 3.7 flash is worse than GPT 5.6 Luna,<00:07:06.560> which<00:07:06.800> is<00:07:06.880> the<00:07:07.039> model<00:07:07.280> over<00:07:07.599> here, 00:07:08.150 --> 00:07:08.160 align:start position:0% Luna, which is the model over here, 00:07:08.160 --> 00:07:10.550 align:start position:0% Luna, which is the model over here, which<00:07:08.400> is<00:07:08.560> achieving<00:07:08.960> a<00:07:09.280> higher<00:07:09.599> score<00:07:10.240> on 00:07:10.550 --> 00:07:10.560 align:start position:0% which is achieving a higher score on 00:07:10.560 --> 00:07:13.189 align:start position:0% which is achieving a higher score on this<00:07:10.800> benchmark<00:07:11.440> deepware<00:07:12.080> engineering<00:07:12.880> for 00:07:13.189 --> 00:07:13.199 align:start position:0% this benchmark deepware engineering for 00:07:13.199 --> 00:07:15.510 align:start position:0% this benchmark deepware engineering for about<00:07:13.520> three<00:07:13.759> times<00:07:14.080> the<00:07:14.400> cost.<00:07:15.039> So,<00:07:15.280> yeah, 00:07:15.510 --> 00:07:15.520 align:start position:0% about three times the cost. So, yeah, 00:07:15.520 --> 00:07:17.029 align:start position:0% about three times the cost. So, yeah, this<00:07:15.759> is<00:07:15.919> kind<00:07:16.080> of<00:07:16.240> interesting<00:07:16.720> because 00:07:17.029 --> 00:07:17.039 align:start position:0% this is kind of interesting because 00:07:17.039 --> 00:07:19.589 align:start position:0% this is kind of interesting because yeah,<00:07:17.199> the<00:07:17.440> cost<00:07:17.599> for<00:07:17.680> Gemini<00:07:18.160> 3.7<00:07:18.720> Flash<00:07:19.120> is<00:07:19.440> a 00:07:19.589 --> 00:07:19.599 align:start position:0% yeah, the cost for Gemini 3.7 Flash is a 00:07:19.599 --> 00:07:21.430 align:start position:0% yeah, the cost for Gemini 3.7 Flash is a little<00:07:19.759> bit<00:07:19.840> more<00:07:20.080> than<00:07:20.319> what<00:07:20.479> it<00:07:20.639> looks<00:07:20.880> like. 00:07:21.430 --> 00:07:21.440 align:start position:0% little bit more than what it looks like. 00:07:21.440 --> 00:07:23.350 align:start position:0% little bit more than what it looks like. Now,<00:07:21.759> some<00:07:22.000> people<00:07:22.240> are<00:07:22.479> a<00:07:22.720> little<00:07:22.800> bit<00:07:22.960> upset 00:07:23.350 --> 00:07:23.360 align:start position:0% Now, some people are a little bit upset 00:07:23.360 --> 00:07:25.110 align:start position:0% Now, some people are a little bit upset and<00:07:23.599> they're<00:07:23.759> like,<00:07:23.919> "Oh,<00:07:24.240> disgraceful. 00:07:25.110 --> 00:07:25.120 align:start position:0% and they're like, "Oh, disgraceful. 00:07:25.120 --> 00:07:27.510 align:start position:0% and they're like, "Oh, disgraceful. Google<00:07:25.440> left<00:07:25.680> out<00:07:25.919> soul,<00:07:26.319> opus,<00:07:26.639> and<00:07:26.880> fable 00:07:27.510 --> 00:07:27.520 align:start position:0% Google left out soul, opus, and fable 00:07:27.520 --> 00:07:29.670 align:start position:0% Google left out soul, opus, and fable because<00:07:27.919> Google<00:07:28.160> is<00:07:28.400> incredibly<00:07:28.880> behind.<00:07:29.599> But 00:07:29.670 --> 00:07:29.680 align:start position:0% because Google is incredibly behind. But 00:07:29.680 --> 00:07:31.029 align:start position:0% because Google is incredibly behind. But I<00:07:29.840> think<00:07:30.000> one<00:07:30.160> thing<00:07:30.240> we<00:07:30.479> got<00:07:30.560> to<00:07:30.720> remember, 00:07:31.029 --> 00:07:31.039 align:start position:0% I think one thing we got to remember, 00:07:31.039 --> 00:07:33.110 align:start position:0% I think one thing we got to remember, guys,<00:07:31.280> is<00:07:31.520> that<00:07:31.759> this<00:07:32.080> model<00:07:32.400> is<00:07:32.560> a<00:07:32.720> flash 00:07:33.110 --> 00:07:33.120 align:start position:0% guys, is that this model is a flash 00:07:33.120 --> 00:07:35.909 align:start position:0% guys, is that this model is a flash model.<00:07:33.520> It's<00:07:33.759> not<00:07:34.000> trying<00:07:34.319> to<00:07:34.479> be<00:07:35.120> a<00:07:35.360> pro<00:07:35.680> model 00:07:35.909 --> 00:07:35.919 align:start position:0% model. It's not trying to be a pro model 00:07:35.919 --> 00:07:37.990 align:start position:0% model. It's not trying to be a pro model or<00:07:36.160> it's<00:07:36.400> not<00:07:36.560> trying<00:07:36.720> to<00:07:36.880> compete<00:07:37.199> with<00:07:37.440> Opus 00:07:37.990 --> 00:07:38.000 align:start position:0% or it's not trying to compete with Opus 00:07:38.000 --> 00:07:40.629 align:start position:0% or it's not trying to compete with Opus or<00:07:38.240> Soul<00:07:38.639> or<00:07:38.800> Fable<00:07:39.280> 5<00:07:39.440> level<00:07:39.759> models<00:07:40.080> yet. 00:07:40.629 --> 00:07:40.639 align:start position:0% or Soul or Fable 5 level models yet. 00:07:40.639 --> 00:07:42.790 align:start position:0% or Soul or Fable 5 level models yet. That<00:07:40.880> is<00:07:41.039> probably<00:07:41.360> going<00:07:41.520> to<00:07:41.599> be<00:07:41.759> Gemini<00:07:42.319> 4. 00:07:42.790 --> 00:07:42.800 align:start position:0% That is probably going to be Gemini 4. 00:07:42.800 --> 00:07:44.870 align:start position:0% That is probably going to be Gemini 4. So<00:07:43.039> when<00:07:43.280> Gemini<00:07:43.759> 4<00:07:44.000> comes<00:07:44.160> out,<00:07:44.400> then<00:07:44.639> I<00:07:44.800> think 00:07:44.870 --> 00:07:44.880 align:start position:0% So when Gemini 4 comes out, then I think 00:07:44.880 --> 00:07:47.189 align:start position:0% So when Gemini 4 comes out, then I think it's<00:07:45.120> okay<00:07:45.360> for<00:07:45.599> us<00:07:45.680> to<00:07:45.919> criticize<00:07:46.400> them<00:07:46.880> if 00:07:47.189 --> 00:07:47.199 align:start position:0% it's okay for us to criticize them if 00:07:47.199 --> 00:07:49.430 align:start position:0% it's okay for us to criticize them if they<00:07:47.360> don't<00:07:47.599> include<00:07:48.080> Soul,<00:07:48.479> Opus,<00:07:48.880> and<00:07:49.039> Fable 00:07:49.430 --> 00:07:49.440 align:start position:0% they don't include Soul, Opus, and Fable 00:07:49.440 --> 00:07:51.189 align:start position:0% they don't include Soul, Opus, and Fable in<00:07:49.599> their<00:07:49.759> benchmark<00:07:50.240> charts<00:07:50.639> because<00:07:50.960> for 00:07:51.189 --> 00:07:51.199 align:start position:0% in their benchmark charts because for 00:07:51.199 --> 00:07:53.189 align:start position:0% in their benchmark charts because for now,<00:07:51.759> I<00:07:52.000> think<00:07:52.080> what<00:07:52.319> they<00:07:52.560> have<00:07:52.720> done<00:07:52.960> is 00:07:53.189 --> 00:07:53.199 align:start position:0% now, I think what they have done is 00:07:53.199 --> 00:07:55.110 align:start position:0% now, I think what they have done is pretty<00:07:53.440> accurate.<00:07:54.160> One<00:07:54.400> lab<00:07:54.639> that<00:07:54.800> I<00:07:54.960> would 00:07:55.110 --> 00:07:55.120 align:start position:0% pretty accurate. One lab that I would 00:07:55.120 --> 00:07:56.629 align:start position:0% pretty accurate. One lab that I would have<00:07:55.280> liked<00:07:55.360> to<00:07:55.520> see<00:07:55.680> or<00:07:55.919> one<00:07:56.160> model<00:07:56.479> for 00:07:56.629 --> 00:07:56.639 align:start position:0% have liked to see or one model for 00:07:56.639 --> 00:07:57.990 align:start position:0% have liked to see or one model for example<00:07:56.960> I<00:07:57.120> would<00:07:57.280> have<00:07:57.360> liked<00:07:57.520> to<00:07:57.680> see<00:07:57.759> on 00:07:57.990 --> 00:07:58.000 align:start position:0% example I would have liked to see on 00:07:58.000 --> 00:07:59.830 align:start position:0% example I would have liked to see on that<00:07:58.160> chart<00:07:58.400> would<00:07:58.639> be<00:07:58.800> Deepseek<00:07:59.360> version<00:07:59.599> 4 00:07:59.830 --> 00:07:59.840 align:start position:0% that chart would be Deepseek version 4 00:07:59.840 --> 00:08:01.510 align:start position:0% that chart would be Deepseek version 4 Flash<00:08:00.319> because<00:08:00.479> that<00:08:00.720> would<00:08:00.800> kind<00:08:00.960> of<00:08:01.120> spoil 00:08:01.510 --> 00:08:01.520 align:start position:0% Flash because that would kind of spoil 00:08:01.520 --> 00:08:03.510 align:start position:0% Flash because that would kind of spoil their<00:08:01.759> release<00:08:02.080> because<00:08:02.319> that<00:08:02.560> model<00:08:02.800> is<00:08:03.120> way 00:08:03.510 --> 00:08:03.520 align:start position:0% their release because that model is way 00:08:03.520 --> 00:08:06.309 align:start position:0% their release because that model is way cheaper<00:08:04.080> compared<00:08:04.479> to<00:08:04.720> the<00:08:04.879> Gemini<00:08:05.360> 3.7<00:08:06.000> flash 00:08:06.309 --> 00:08:06.319 align:start position:0% cheaper compared to the Gemini 3.7 flash 00:08:06.319 --> 00:08:08.790 align:start position:0% cheaper compared to the Gemini 3.7 flash model<00:08:06.560> lineup.<00:08:07.360> Now<00:08:07.680> this<00:08:07.840> model<00:08:08.240> jumped<00:08:08.560> from 00:08:08.790 --> 00:08:08.800 align:start position:0% model lineup. Now this model jumped from 00:08:08.800 --> 00:08:11.189 align:start position:0% model lineup. Now this model jumped from number<00:08:09.120> 19<00:08:09.440> to<00:08:09.759> 8<00:08:10.000> on<00:08:10.240> the<00:08:10.479> web<00:08:10.800> development 00:08:11.189 --> 00:08:11.199 align:start position:0% number 19 to 8 on the web development 00:08:11.199 --> 00:08:13.589 align:start position:0% number 19 to 8 on the web development area<00:08:11.440> and<00:08:11.680> we<00:08:11.840> see<00:08:12.080> it<00:08:12.240> over<00:08:12.479> here<00:08:12.800> now<00:08:13.280> and 00:08:13.589 --> 00:08:13.599 align:start position:0% area and we see it over here now and 00:08:13.599 --> 00:08:15.110 align:start position:0% area and we see it over here now and couple<00:08:13.759> of<00:08:13.919> models<00:08:14.319> that<00:08:14.400> are<00:08:14.560> ahead<00:08:14.800> of<00:08:14.879> it 00:08:15.110 --> 00:08:15.120 align:start position:0% couple of models that are ahead of it 00:08:15.120 --> 00:08:18.150 align:start position:0% couple of models that are ahead of it are<00:08:15.280> Opus<00:08:15.759> 5<00:08:16.000> obviously<00:08:16.319> Kim<00:08:16.720> K3<00:08:17.199> Quinn<00:08:17.599> 3.8 00:08:18.150 --> 00:08:18.160 align:start position:0% are Opus 5 obviously Kim K3 Quinn 3.8 00:08:18.160 --> 00:08:21.670 align:start position:0% are Opus 5 obviously Kim K3 Quinn 3.8 Max<00:08:18.639> Cloud<00:08:18.960> Opus<00:08:19.360> 5<00:08:20.000> Gro<00:08:20.479> 4.6<00:08:20.960> 6,<00:08:21.280> which<00:08:21.440> is<00:08:21.520> a 00:08:21.670 --> 00:08:21.680 align:start position:0% Max Cloud Opus 5 Gro 4.6 6, which is a 00:08:21.680 --> 00:08:23.350 align:start position:0% Max Cloud Opus 5 Gro 4.6 6, which is a model<00:08:21.919> that<00:08:22.080> came<00:08:22.319> yesterday,<00:08:22.800> which<00:08:23.039> was<00:08:23.199> a 00:08:23.350 --> 00:08:23.360 align:start position:0% model that came yesterday, which was a 00:08:23.360 --> 00:08:26.710 align:start position:0% model that came yesterday, which was a big<00:08:23.520> win<00:08:23.759> for<00:08:24.000> SpaceX,<00:08:24.879> Fable<00:08:25.360> 5,<00:08:25.599> and<00:08:25.840> 5.6. 00:08:26.710 --> 00:08:26.720 align:start position:0% big win for SpaceX, Fable 5, and 5.6. 00:08:26.720 --> 00:08:28.950 align:start position:0% big win for SpaceX, Fable 5, and 5.6. Soul.<00:08:27.360> So,<00:08:27.599> yeah,<00:08:27.840> this<00:08:28.000> model<00:08:28.319> is<00:08:28.639> trying<00:08:28.800> to 00:08:28.950 --> 00:08:28.960 align:start position:0% Soul. So, yeah, this model is trying to 00:08:28.960 --> 00:08:31.110 align:start position:0% Soul. So, yeah, this model is trying to compete<00:08:29.360> in<00:08:29.520> the<00:08:29.680> web<00:08:29.919> development,<00:08:30.639> but<00:08:30.879> it's 00:08:31.110 --> 00:08:31.120 align:start position:0% compete in the web development, but it's 00:08:31.120 --> 00:08:33.110 align:start position:0% compete in the web development, but it's still<00:08:31.280> behind<00:08:31.520> all<00:08:31.680> of<00:08:31.840> these<00:08:32.080> models,<00:08:32.880> which 00:08:33.110 --> 00:08:33.120 align:start position:0% still behind all of these models, which 00:08:33.120 --> 00:08:35.110 align:start position:0% still behind all of these models, which is,<00:08:33.360> you<00:08:33.519> know,<00:08:33.760> expected<00:08:34.240> cuz<00:08:34.479> it's<00:08:34.640> a<00:08:34.800> flash 00:08:35.110 --> 00:08:35.120 align:start position:0% is, you know, expected cuz it's a flash 00:08:35.120 --> 00:08:37.269 align:start position:0% is, you know, expected cuz it's a flash model.<00:08:35.519> It's<00:08:35.760> not<00:08:35.839> really<00:08:36.080> a<00:08:36.240> pro<00:08:36.560> model. 00:08:37.269 --> 00:08:37.279 align:start position:0% model. It's not really a pro model. 00:08:37.279 --> 00:08:39.589 align:start position:0% model. It's not really a pro model. Today,<00:08:37.599> OpenAI<00:08:38.240> has<00:08:38.479> also<00:08:38.800> launched<00:08:39.120> a<00:08:39.279> weight 00:08:39.589 --> 00:08:39.599 align:start position:0% Today, OpenAI has also launched a weight 00:08:39.599 --> 00:08:42.469 align:start position:0% Today, OpenAI has also launched a weight list<00:08:39.760> for<00:08:40.080> 5.6<00:08:40.880> so<00:08:41.279> ultra<00:08:41.599> fast<00:08:41.839> mode.<00:08:42.320> Now, 00:08:42.469 --> 00:08:42.479 align:start position:0% list for 5.6 so ultra fast mode. Now, 00:08:42.479 --> 00:08:44.949 align:start position:0% list for 5.6 so ultra fast mode. Now, this<00:08:42.719> is<00:08:42.880> possible<00:08:43.360> because<00:08:44.080> of<00:08:44.320> the<00:08:44.560> fast 00:08:44.949 --> 00:08:44.959 align:start position:0% this is possible because of the fast 00:08:44.959 --> 00:08:46.630 align:start position:0% this is possible because of the fast chips<00:08:45.279> that<00:08:45.440> they<00:08:45.680> have<00:08:45.839> access<00:08:46.160> to<00:08:46.399> now 00:08:46.630 --> 00:08:46.640 align:start position:0% chips that they have access to now 00:08:46.640 --> 00:08:49.829 align:start position:0% chips that they have access to now through<00:08:46.959> Cabus<00:08:47.839> and<00:08:48.000> GPT<00:08:48.560> 5.6<00:08:48.959> 6o<00:08:49.519> with<00:08:49.680> this 00:08:49.829 --> 00:08:49.839 align:start position:0% through Cabus and GPT 5.6 6o with this 00:08:49.839 --> 00:08:52.150 align:start position:0% through Cabus and GPT 5.6 6o with this chip<00:08:50.240> kind<00:08:50.480> of<00:08:50.640> running<00:08:50.880> it<00:08:51.519> is<00:08:51.839> able<00:08:52.000> to 00:08:52.150 --> 00:08:52.160 align:start position:0% chip kind of running it is able to 00:08:52.160 --> 00:08:53.990 align:start position:0% chip kind of running it is able to achieve<00:08:52.480> an<00:08:52.800> ultra<00:08:53.120> fast<00:08:53.440> mode<00:08:53.680> that 00:08:53.990 --> 00:08:54.000 align:start position:0% achieve an ultra fast mode that 00:08:54.000 --> 00:08:57.110 align:start position:0% achieve an ultra fast mode that generates<00:08:54.399> up<00:08:54.640> to<00:08:54.959> 750<00:08:55.920> output<00:08:56.399> tokens<00:08:56.800> per 00:08:57.110 --> 00:08:57.120 align:start position:0% generates up to 750 output tokens per 00:08:57.120 --> 00:09:00.389 align:start position:0% generates up to 750 output tokens per second<00:08:57.920> which<00:08:58.160> is<00:08:58.480> about<00:08:58.959> 14<00:08:59.440> times<00:08:59.920> faster 00:09:00.389 --> 00:09:00.399 align:start position:0% second which is about 14 times faster 00:09:00.399 --> 00:09:02.389 align:start position:0% second which is about 14 times faster than<00:09:00.640> the<00:09:00.880> standard<00:09:01.200> mode.<00:09:01.760> So<00:09:01.920> we're<00:09:02.160> getting 00:09:02.389 --> 00:09:02.399 align:start position:0% than the standard mode. So we're getting 00:09:02.399 --> 00:09:06.310 align:start position:0% than the standard mode. So we're getting a<00:09:02.880> really<00:09:03.200> fast<00:09:03.519> version<00:09:03.760> of<00:09:04.000> GPT<00:09:04.560> 5.6<00:09:05.279> so<00:09:06.080> now 00:09:06.310 --> 00:09:06.320 align:start position:0% a really fast version of GPT 5.6 so now 00:09:06.320 --> 00:09:08.870 align:start position:0% a really fast version of GPT 5.6 so now this<00:09:06.480> is<00:09:06.720> supposed<00:09:06.959> to<00:09:07.120> be<00:09:07.279> used<00:09:07.600> for<00:09:08.240> live<00:09:08.560> or 00:09:08.870 --> 00:09:08.880 align:start position:0% this is supposed to be used for live or 00:09:08.880 --> 00:09:10.710 align:start position:0% this is supposed to be used for live or near<00:09:09.200> production<00:09:09.680> workloads<00:09:10.240> like<00:09:10.480> you<00:09:10.560> know 00:09:10.710 --> 00:09:10.720 align:start position:0% near production workloads like you know 00:09:10.720 --> 00:09:13.269 align:start position:0% near production workloads like you know real-time<00:09:11.279> voice<00:09:11.680> support<00:09:12.240> commerce<00:09:13.040> what 00:09:13.269 --> 00:09:13.279 align:start position:0% real-time voice support commerce what 00:09:13.279 --> 00:09:15.829 align:start position:0% real-time voice support commerce what this<00:09:13.440> allows<00:09:13.839> GPT<00:09:14.399> 5.6<00:09:14.880> six<00:09:15.120> soul<00:09:15.360> to<00:09:15.519> do<00:09:15.600> is 00:09:15.829 --> 00:09:15.839 align:start position:0% this allows GPT 5.6 six soul to do is 00:09:15.839 --> 00:09:17.670 align:start position:0% this allows GPT 5.6 six soul to do is kind<00:09:16.000> of<00:09:16.160> be<00:09:16.320> really<00:09:16.640> fast<00:09:16.959> in<00:09:17.279> critical 00:09:17.670 --> 00:09:17.680 align:start position:0% kind of be really fast in critical 00:09:17.680 --> 00:09:19.509 align:start position:0% kind of be really fast in critical situations<00:09:18.399> when<00:09:18.800> people<00:09:19.120> might<00:09:19.360> be 00:09:19.509 --> 00:09:19.519 align:start position:0% situations when people might be 00:09:19.519 --> 00:09:21.670 align:start position:0% situations when people might be interacting<00:09:20.000> with<00:09:20.240> the<00:09:20.399> AI<00:09:20.720> agent<00:09:21.440> like 00:09:21.670 --> 00:09:21.680 align:start position:0% interacting with the AI agent like 00:09:21.680 --> 00:09:24.630 align:start position:0% interacting with the AI agent like financial<00:09:22.240> research<00:09:22.880> security<00:09:23.279> response 00:09:24.630 --> 00:09:24.640 align:start position:0% financial research security response 00:09:24.640 --> 00:09:26.470 align:start position:0% financial research security response support<00:09:25.120> I<00:09:25.360> think<00:09:25.440> is<00:09:25.680> going<00:09:25.839> to<00:09:26.000> be<00:09:26.080> a<00:09:26.320> big 00:09:26.470 --> 00:09:26.480 align:start position:0% support I think is going to be a big 00:09:26.480 --> 00:09:29.350 align:start position:0% support I think is going to be a big area<00:09:26.800> where<00:09:27.440> this<00:09:27.680> new<00:09:28.080> ultraast<00:09:28.640> mode<00:09:29.120> will 00:09:29.350 --> 00:09:29.360 align:start position:0% area where this new ultraast mode will 00:09:29.360 --> 00:09:31.990 align:start position:0% area where this new ultraast mode will be<00:09:29.519> kind<00:09:29.760> of<00:09:29.920> implemented<00:09:31.120> business<00:09:31.600> and 00:09:31.990 --> 00:09:32.000 align:start position:0% be kind of implemented business and 00:09:32.000 --> 00:09:33.910 align:start position:0% be kind of implemented business and developer<00:09:32.399> agents<00:09:32.880> maybe<00:09:33.200> yeah<00:09:33.440> but<00:09:33.680> I<00:09:33.839> think 00:09:33.910 --> 00:09:33.920 align:start position:0% developer agents maybe yeah but I think 00:09:33.920 --> 00:09:35.590 align:start position:0% developer agents maybe yeah but I think like<00:09:34.240> support<00:09:34.560> or<00:09:34.800> near<00:09:35.200> production 00:09:35.590 --> 00:09:35.600 align:start position:0% like support or near production 00:09:35.600 --> 00:09:37.750 align:start position:0% like support or near production workloads<00:09:36.160> like<00:09:36.320> real-time<00:09:36.880> voice<00:09:37.440> I<00:09:37.680> see 00:09:37.750 --> 00:09:37.760 align:start position:0% workloads like real-time voice I see 00:09:37.760 --> 00:09:40.150 align:start position:0% workloads like real-time voice I see this<00:09:38.000> model<00:09:38.320> really<00:09:38.640> excelling<00:09:39.040> at<00:09:39.279> that<00:09:39.920> and 00:09:40.150 --> 00:09:40.160 align:start position:0% this model really excelling at that and 00:09:40.160 --> 00:09:42.150 align:start position:0% this model really excelling at that and obviously<00:09:40.640> this<00:09:40.880> is<00:09:41.120> still<00:09:41.360> a<00:09:41.680> weightless 00:09:42.150 --> 00:09:42.160 align:start position:0% obviously this is still a weightless 00:09:42.160 --> 00:09:44.150 align:start position:0% obviously this is still a weightless mode<00:09:42.480> we<00:09:42.720> don't<00:09:42.880> know<00:09:43.360> how<00:09:43.600> many<00:09:43.760> people<00:09:44.000> are 00:09:44.150 --> 00:09:44.160 align:start position:0% mode we don't know how many people are 00:09:44.160 --> 00:09:45.750 align:start position:0% mode we don't know how many people are going<00:09:44.240> to<00:09:44.399> get<00:09:44.640> access<00:09:44.959> to<00:09:45.120> this,<00:09:45.440> how 00:09:45.750 --> 00:09:45.760 align:start position:0% going to get access to this, how 00:09:45.760 --> 00:09:47.910 align:start position:0% going to get access to this, how successful<00:09:46.240> it<00:09:46.480> is<00:09:46.640> or<00:09:46.880> what<00:09:47.040> the<00:09:47.279> pricing<00:09:47.600> is. 00:09:47.910 --> 00:09:47.920 align:start position:0% successful it is or what the pricing is. 00:09:47.920 --> 00:09:49.430 align:start position:0% successful it is or what the pricing is. I<00:09:48.080> don't<00:09:48.160> know<00:09:48.240> if<00:09:48.399> the<00:09:48.560> pricing<00:09:48.959> has<00:09:49.120> changed 00:09:49.430 --> 00:09:49.440 align:start position:0% I don't know if the pricing has changed 00:09:49.440 --> 00:09:51.190 align:start position:0% I don't know if the pricing has changed because<00:09:49.680> there's<00:09:50.000> no<00:09:50.240> information<00:09:50.640> in<00:09:50.959> their 00:09:51.190 --> 00:09:51.200 align:start position:0% because there's no information in their 00:09:51.200 --> 00:09:53.990 align:start position:0% because there's no information in their actual,<00:09:52.160> you<00:09:52.320> know,<00:09:52.480> blog<00:09:52.800> post.<00:09:53.440> But<00:09:53.600> as<00:09:53.839> I 00:09:53.990 --> 00:09:54.000 align:start position:0% actual, you know, blog post. But as I 00:09:54.000 --> 00:09:55.910 align:start position:0% actual, you know, blog post. But as I mentioned,<00:09:54.480> couple<00:09:54.720> of<00:09:54.880> areas<00:09:55.279> where<00:09:55.600> they 00:09:55.910 --> 00:09:55.920 align:start position:0% mentioned, couple of areas where they 00:09:55.920 --> 00:09:57.350 align:start position:0% mentioned, couple of areas where they mentioned<00:09:56.240> that<00:09:56.560> this<00:09:56.720> is<00:09:56.880> going<00:09:57.040> to<00:09:57.200> be 00:09:57.350 --> 00:09:57.360 align:start position:0% mentioned that this is going to be 00:09:57.360 --> 00:09:59.350 align:start position:0% mentioned that this is going to be really<00:09:57.600> important.<00:09:58.399> Customer<00:09:58.880> support<00:09:59.200> and 00:09:59.350 --> 00:09:59.360 align:start position:0% really important. Customer support and 00:09:59.360 --> 00:10:01.190 align:start position:0% really important. Customer support and voice,<00:09:59.760> commerce,<00:10:00.320> live<00:10:00.640> research<00:10:00.959> and 00:10:01.190 --> 00:10:01.200 align:start position:0% voice, commerce, live research and 00:10:01.200 --> 00:10:03.750 align:start position:0% voice, commerce, live research and experimentation,<00:10:02.480> financial<00:10:03.120> research<00:10:03.440> and 00:10:03.750 --> 00:10:03.760 align:start position:0% experimentation, financial research and 00:10:03.760 --> 00:10:05.829 align:start position:0% experimentation, financial research and security,<00:10:04.640> incident<00:10:05.200> response<00:10:05.600> and 00:10:05.829 --> 00:10:05.839 align:start position:0% security, incident response and 00:10:05.839 --> 00:10:07.750 align:start position:0% security, incident response and reliability.<00:10:06.880> But<00:10:07.040> yeah,<00:10:07.200> I<00:10:07.440> think<00:10:07.519> this 00:10:07.750 --> 00:10:07.760 align:start position:0% reliability. But yeah, I think this 00:10:07.760 --> 00:10:09.670 align:start position:0% reliability. But yeah, I think this partnership<00:10:08.320> is<00:10:08.560> going<00:10:08.720> to<00:10:08.880> be<00:10:09.040> important<00:10:09.440> for 00:10:09.670 --> 00:10:09.680 align:start position:0% partnership is going to be important for 00:10:09.680 --> 00:10:12.230 align:start position:0% partnership is going to be important for OpenAI<00:10:10.320> going<00:10:10.560> forward.<00:10:11.200> But<00:10:11.360> as<00:10:11.519> I<00:10:11.680> said,<00:10:11.920> 14 00:10:12.230 --> 00:10:12.240 align:start position:0% OpenAI going forward. But as I said, 14 00:10:12.240 --> 00:10:14.230 align:start position:0% OpenAI going forward. But as I said, 14 times<00:10:12.560> the<00:10:12.800> speed.<00:10:13.120> What<00:10:13.360> does<00:10:13.519> that<00:10:13.760> mean<00:10:13.920> for 00:10:14.230 --> 00:10:14.240 align:start position:0% times the speed. What does that mean for 00:10:14.240 --> 00:10:16.790 align:start position:0% times the speed. What does that mean for cost?<00:10:14.720> We<00:10:14.880> don't<00:10:15.040> know<00:10:15.200> yet.<00:10:16.240> But<00:10:16.399> that's<00:10:16.640> it 00:10:16.790 --> 00:10:16.800 align:start position:0% cost? We don't know yet. But that's it 00:10:16.800 --> 00:10:18.550 align:start position:0% cost? We don't know yet. But that's it for<00:10:17.040> today's<00:10:17.360> video.<00:10:17.920> Make<00:10:18.079> sure<00:10:18.240> you<00:10:18.399> guys 00:10:18.550 --> 00:10:18.560 align:start position:0% for today's video. Make sure you guys 00:10:18.560 --> 00:10:20.470 align:start position:0% for today's video. Make sure you guys are<00:10:18.720> subscribed<00:10:19.120> to<00:10:19.279> the<00:10:19.519> channel.<00:10:20.160> Follow 00:10:20.470 --> 00:10:20.480 align:start position:0% are subscribed to the channel. Follow 00:10:20.480 --> 00:10:22.150 align:start position:0% are subscribed to the channel. Follow our<00:10:20.800> new<00:10:20.959> newsletter<00:10:21.519> as<00:10:21.760> well<00:10:21.920> at 00:10:22.150 --> 00:10:22.160 align:start position:0% our new newsletter as well at 00:10:22.160 --> 00:10:24.389 align:start position:0% our new newsletter as well at universeai.behive.com 00:10:24.389 --> 00:10:24.399 align:start position:0% universeai.behive.com 00:10:24.399 --> 00:10:26.069 align:start position:0% universeai.behive.com as<00:10:24.640> well<00:10:24.720> as<00:10:24.880> subscribe<00:10:25.200> to<00:10:25.360> the<00:10:25.519> main<00:10:25.760> channel 00:10:26.069 --> 00:10:26.079 align:start position:0% as well as subscribe to the main channel 00:10:26.079 --> 00:10:28.550 align:start position:0% as well as subscribe to the main channel World<00:10:26.240> of<00:10:26.480> AI<00:10:27.120> and<00:10:27.360> support<00:10:27.680> us<00:10:27.839> on<00:10:28.079> X<00:10:28.399> by 00:10:28.550 --> 00:10:28.560 align:start position:0% World of AI and support us on X by 00:10:28.560 --> 00:10:30.790 align:start position:0% World of AI and support us on X by following<00:10:28.800> the<00:10:29.040> Universe<00:10:29.360> of<00:10:29.519> AIZ<00:10:30.160> as<00:10:30.320> well. 00:10:30.790 --> 00:10:30.800 align:start position:0% following the Universe of AIZ as well. 00:10:30.800 --> 00:10:32.389 align:start position:0% following the Universe of AIZ as well. Until<00:10:31.120> then,<00:10:31.440> I'll<00:10:31.600> see<00:10:31.760> you<00:10:31.920> guys<00:10:32.079> in<00:10:32.240> the 00:10:32.389 --> 00:10:32.399 align:start position:0% Until then, I'll see you guys in the 00:10:32.399 --> 00:10:34.560 align:start position:0% Until then, I'll see you guys in the next