1
00:00:00,000 --> 00:00:02,280
I believe that was something I've seen.

2
00:00:05,220 --> 00:00:08,380
The trees are quite reminiscent of a poo.

3
00:00:08,660 --> 00:00:12,800
DeepSeek V4 Pro has just been released in general available version,

4
00:00:12,960 --> 00:00:18,800
which also happened with the V4 Flash model back about 12 or 13 days ago on 0731.

5
00:00:19,260 --> 00:00:24,260
Now that model, when tested, proved to be a significant, significant increase in capability

6
00:00:24,260 --> 00:00:26,940
over the preview version that was previously around.

7
00:00:26,940 --> 00:00:31,660
and today we're going to be finding out if the same can be said for the newly updated Checkpoint

8
00:00:31,660 --> 00:00:35,760
general availability for V4 Pro dubbed 0813.

9
00:00:35,760 --> 00:00:40,660
So before we get into it, do feel free to subscribe so we can hit 70k and then 100k with the plaque.

10
00:00:41,020 --> 00:00:44,140
And with that, basically there's not a lot to say about this right now

11
00:00:44,140 --> 00:00:46,820
because it seems to have been more or less silently released.

12
00:00:46,980 --> 00:00:52,020
There's no official announcement post yet on X, which I do believe is generally where we would see this,

13
00:00:52,360 --> 00:00:55,940
but they do mention here just in the website a couple references to it

14
00:00:55,940 --> 00:01:02,940
如果使用这个设计划来这里,你会用这个设计划来到新的V4Pro0813。

15
00:01:02,940 --> 00:01:32,940
另外一方,它是在开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放开放

16
00:01:32,940 --> 00:01:37,720
selectable thinking modes as you can see if we do click on this so it's possible to kind of tone

17
00:01:37,720 --> 00:01:42,200
that down but in the interest of seeing the maximum potential performance all of our coding

18
00:01:42,200 --> 00:01:48,220
tasks for today are going to be run on this maximum effort now i want to mention the following with a

19
00:01:48,220 --> 00:01:52,480
grain of salt because i cannot definitively confirm that this is ground truth but i would

20
00:01:52,480 --> 00:01:57,600
imagine it probably is there were some references and photos floating around which apparently came

21
00:01:57,600 --> 00:02:02,520
from a group related to deep seek i believe they said it was a we chat group about deep seek or

22
00:02:02,520 --> 00:02:12,380
and if this is to be believed we can see that these would hypothetically be the benchmark results for DeepSeek V4 Pro 0813 or the general availability model.

23
00:02:12,600 --> 00:02:18,280
And basically my main interest here is just seeing the leaping capability over the V4 Pro preview to now

24
00:02:18,280 --> 00:02:24,320
because I was initially disappointed with the V4 Pro preview as well as the V4 Flash preview.

25
00:02:24,740 --> 00:02:30,200
They felt very undercooked and probably because they were named preview that's actually like fair it makes sense.

26
00:02:30,200 --> 00:02:35,260
but again i want to reiterate the leaping capability between flash and then the 0731

27
00:02:35,260 --> 00:02:40,020
general availability version of flash was absolutely massive this is like the best

28
00:02:40,020 --> 00:02:46,380
local model right now for like a dual spark setup so i mean in a niche you know but still

29
00:02:46,380 --> 00:02:51,920
if that leaping capability remains then what we can see right here just in this deep swe score

30
00:02:51,920 --> 00:02:58,680
increase from 12.8 to 62.7 would then hypothetically put this above things like opus 4.8

31
00:02:58,680 --> 00:03:03,940
and make it very very competitive being especially that this is an open weight model or will be

32
00:03:03,940 --> 00:03:09,820
that's also extremely exciting because this is not like horrifically gigantic i guess when

33
00:03:09,820 --> 00:03:16,120
considering things like kimmy k3 or quen 3.8 max it's still big at 1.6 trillion parameters it is

34
00:03:16,120 --> 00:03:21,480
a mixture of experts model with how many active so if we do estimate that this is going to be the

35
00:03:21,480 --> 00:03:27,080
same size as the preview version of v4 pro it's a 1.6 trillion parameter mixture of experts model

36
00:03:27,080 --> 00:03:33,180
with 49 billion active so it is still very very large but it's not as gigantic as like the

37
00:03:33,180 --> 00:03:38,220
closing in on 3 trillion parameter models so additionally to that pricing and this is something

38
00:03:38,220 --> 00:03:43,140
that we need to basically take just as it is right now because they mention and have been for quite

39
00:03:43,140 --> 00:03:48,340
a bit of time that their api pricing is going to be significantly increased soon now significantly

40
00:03:48,340 --> 00:03:54,280
is relative to what it currently is which is like very very very cheap so right now pricing per

41
00:03:54,280 --> 00:04:03,700
million input tokens for DeepSeek V4 Pro is 4.35 cents so okay no I know what this is but like last

42
00:04:03,700 --> 00:04:08,860
time I messed up on this one and I said it was two cents and it wasn't and then I got roasted so

43
00:04:08,860 --> 00:04:15,900
this is zero 43 cents per million input tokens and 87 cents per million output tokens is the

44
00:04:15,900 --> 00:04:21,280
current price now if this gets like a significant increase say it goes up like five or ten times it

45
00:04:21,280 --> 00:04:51,260
优优优优独播剧场——YoYo Television Series Exclusive

46
00:04:51,280 --> 00:04:54,900
so I also have it routed and working through this desktop app as well

47
00:04:54,900 --> 00:04:57,260
and it is doing a skate game for us in C++

48
00:04:57,260 --> 00:04:59,380
I would imagine that will take quite a bit of time

49
00:04:59,380 --> 00:05:03,560
so I did just want to start it and verify this app was actually working properly on this system

50
00:05:03,560 --> 00:05:04,680
and it does seem to

51
00:05:04,680 --> 00:05:08,040
so keep in mind any mention of GPT or Codex here

52
00:05:08,040 --> 00:05:13,120
is routed directly to DeepSeek on max reasoning effort for the 0813 model

53
00:05:13,120 --> 00:05:15,080
and with that let's take a look at our

54
00:05:15,080 --> 00:05:19,560
okay so apparently the OpenAI beta application

55
00:05:19,560 --> 00:05:23,360
設定 itself to be the default for all file types

56
00:05:23,360 --> 00:05:24,480
that's infuriating

57
00:05:24,480 --> 00:05:28,100
very quen 3.8 aesthetic

58
00:05:28,100 --> 00:05:30,020
i'm going to turn sound on

59
00:05:30,020 --> 00:05:31,700
because i did see some mentions of it

60
00:05:31,700 --> 00:05:32,940
just talking about like

61
00:05:32,940 --> 00:05:34,620
sounds

62
00:05:34,620 --> 00:05:36,380
it starts in our notepad

63
00:05:36,380 --> 00:05:38,400
can we resize this

64
00:05:38,400 --> 00:05:40,280
acceptable

65
00:05:40,280 --> 00:05:40,920
minimize

66
00:05:40,920 --> 00:05:42,080
unminimize

67
00:05:42,080 --> 00:05:42,640
good

68
00:05:42,640 --> 00:05:43,680
the big one

69
00:05:43,680 --> 00:05:44,620
is there a right click

70
00:05:44,620 --> 00:05:46,020
yeah there is

71
00:05:46,020 --> 00:05:52,020
有同一種正確的時間和時間在我的手上,

72
00:05:52,020 --> 00:05:55,020
在我右上的特定是一個特別的特定,

73
00:05:55,020 --> 00:05:57,020
是一個特定的特定,

74
00:05:57,020 --> 00:05:59,020
所以我們會先把這個作品看起來。

75
00:05:59,020 --> 00:06:01,020
我們開始看了這個Files,

76
00:06:01,020 --> 00:06:05,020
我沒有看到一個Files icon像這個樣子,

77
00:06:05,020 --> 00:06:08,020
我們有些比較有的確 transparency

78
00:06:08,020 --> 00:06:10,020
和一個很像SynthWave的 theme,

79
00:06:10,020 --> 00:06:12,020
我個人喜歡的,

80
00:06:12,020 --> 00:06:14,020
所以我們不太喜歡的Files,

81
00:06:14,020 --> 00:06:16,020
我們不可以看過它,

82
00:06:16,020 --> 00:06:18,020
可以可以做新的东西

83
00:06:18,020 --> 00:06:18,740
可以可以

84
00:06:18,740 --> 00:06:20,020
可以可以做新的东西

85
00:06:20,020 --> 00:06:22,020
我们也可以

86
00:06:22,020 --> 00:06:24,020
可以可以做新的东西

87
00:06:24,020 --> 00:06:25,020
可以做新的东西

88
00:06:25,020 --> 00:06:27,020
可以做新的东西

89
00:06:27,020 --> 00:06:28,020
可以做新的东西

90
00:06:28,020 --> 00:06:29,020
然后上

91
00:06:29,020 --> 00:06:31,020
可以做新的东西

92
00:06:31,020 --> 00:06:32,020
不错

93
00:06:32,020 --> 00:06:33,020
不错

94
00:06:33,020 --> 00:06:34,020
不错

95
00:06:34,020 --> 00:06:35,020
在这些什么

96
00:06:35,020 --> 00:06:36,020
在这些什么

97
00:06:36,020 --> 00:06:37,020
是这些小组的

98
00:06:37,020 --> 00:06:38,020
不错

99
00:06:38,020 --> 00:06:39,020
我们也看到了

100
00:06:39,020 --> 00:06:40,020
因为它开了

101
00:06:40,020 --> 00:06:41,020
这些东西

102
00:06:41,020 --> 00:06:42,020
只要给我们一些

103
00:06:42,020 --> 00:06:43,020
我喜欢看

104
00:06:43,020 --> 00:06:44,020
是我们的信号

105
00:06:44,020 --> 00:06:45,020
是我们的信号

106
00:06:45,020 --> 00:06:46,120
No such directory

107
00:06:46,120 --> 00:06:48,840
Didn't I delete the documents directory

108
00:06:48,840 --> 00:06:49,800
Hold on a second

109
00:06:49,800 --> 00:06:50,740
That's okay

110
00:06:50,740 --> 00:06:52,940
Oh it remembered that I did that

111
00:06:52,940 --> 00:06:54,160
I might have messed that up

112
00:06:54,160 --> 00:06:55,080
So that's gonna be

113
00:06:55,080 --> 00:06:55,900
That's on me

114
00:06:55,900 --> 00:06:57,940
Okay this right here

115
00:06:57,940 --> 00:06:59,980
Is just not right

116
00:06:59,980 --> 00:07:03,160
It should not have a like visible text box

117
00:07:03,160 --> 00:07:04,580
That happens sometimes

118
00:07:04,580 --> 00:07:06,560
Help did give us things

119
00:07:06,560 --> 00:07:07,860
Let's just do NeoFetch

120
00:07:07,860 --> 00:07:11,900
That's actually pretty well done right there

121
00:07:14,460 --> 00:07:14,900
Negula

122
00:07:14,900 --> 00:07:16,760
or nebula

123
00:07:16,760 --> 00:07:19,180
still that's

124
00:07:19,180 --> 00:07:21,760
actually alright I'm just being kind of mean

125
00:07:21,760 --> 00:07:22,620
memory

126
00:07:22,620 --> 00:07:23,940
this is fine

127
00:07:23,940 --> 00:07:27,700
it actually put a bit of cleverness in there

128
00:07:27,700 --> 00:07:29,500
1.44 megs

129
00:07:29,500 --> 00:07:30,820
out of infinity this is fine

130
00:07:30,820 --> 00:07:33,360
isn't that how much an old school floppy disk would hold

131
00:07:33,360 --> 00:07:35,660
I don't know because those are older than me

132
00:07:35,660 --> 00:07:36,540
next up

133
00:07:36,540 --> 00:07:39,040
calculator okay so 47

134
00:07:39,040 --> 00:07:40,340
times

135
00:07:40,340 --> 00:07:41,780
8

136
00:07:41,780 --> 00:07:45,240
376

137
00:07:45,240 --> 00:07:47,060
我都没有做一种的一种

138
00:07:47,060 --> 00:07:47,880
分之六

139
00:07:47,880 --> 00:07:48,700
好

140
00:07:48,700 --> 00:07:50,020
設定

141
00:07:50,020 --> 00:07:52,780
一定

142
00:07:52,780 --> 00:07:54,240
seeing a bit of like

143
00:07:54,240 --> 00:07:55,160
Quinn 3.8

144
00:07:55,160 --> 00:07:55,840
in here

145
00:07:55,840 --> 00:07:56,760
just in terms of

146
00:07:56,760 --> 00:07:57,740
some of the style here

147
00:07:57,740 --> 00:07:58,700
what is that

148
00:07:58,700 --> 00:07:59,880
and why do I like it

149
00:08:01,541 --> 00:08:02,541
Aurora。

150
00:08:02,541 --> 00:08:03,541
OK。

151
00:08:03,541 --> 00:08:05,541
City。

152
00:08:05,541 --> 00:08:07,541
這看起來。

153
00:08:07,541 --> 00:08:09,541
這看起來是一種類似的。

154
00:08:09,541 --> 00:08:11,541
這看起來是一種類似的。

155
00:08:11,541 --> 00:08:13,541
這看起來是一種類似的。

156
00:08:13,541 --> 00:08:15,621
它其实看了一 ton

157
00:08:15,621 --> 00:08:17,261
像这个写的写的

158
00:08:17,261 --> 00:08:18,781
我们有的

159
00:08:18,781 --> 00:08:19,541
我们有的

160
00:08:19,541 --> 00:08:20,301
负责的

161
00:08:20,301 --> 00:08:21,301
负责的

162
00:08:21,301 --> 00:08:22,221
负责

163
00:08:22,221 --> 00:08:22,501
负责

164
00:08:22,501 --> 00:08:22,941
负责

165
00:08:22,941 --> 00:08:23,501
负责

166
00:08:23,501 --> 00:08:24,501
负责

167
00:08:24,501 --> 00:08:25,501
负责

168
00:08:25,501 --> 00:08:26,501
负责

169
00:08:26,501 --> 00:08:27,501
负责

170
00:08:27,501 --> 00:08:28,501
负责

171
00:08:28,501 --> 00:08:29,501
负责

172
00:08:29,501 --> 00:08:30,501
负责

173
00:08:30,501 --> 00:08:31,501
责

174
00:08:31,501 --> 00:08:31,961
责

175
00:08:31,961 --> 00:08:32,501
责

176
00:08:32,501 --> 00:08:33,501
责

177
00:08:33,501 --> 00:08:34,501
责

178
00:08:34,501 --> 00:08:35,501
责

179
00:08:35,501 --> 00:08:36,501
责

180
00:08:36,501 --> 00:08:37,501
责

181
00:08:37,501 --> 00:08:39,501
责

182
00:08:39,501 --> 00:08:41,501
责

183
00:08:41,501 --> 00:08:43,501
责

184
00:08:43,541 --> 00:09:13,541
OK,所以你需要按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按按

185
00:09:13,541 --> 00:09:14,541
It's troubled

186
00:09:14,541 --> 00:09:18,701
But

187
00:09:18,701 --> 00:09:23,321
The sounds for like the smack are kind of funny

188
00:09:23,321 --> 00:09:24,781
Alright

189
00:09:24,781 --> 00:09:25,821
And we wrecked

190
00:09:25,821 --> 00:09:28,041
And we only grabbed 100 out of 1000

191
00:09:28,041 --> 00:09:29,481
But it did work

192
00:09:29,481 --> 00:09:32,341
So keep in mind we need to press enter to play these games

193
00:09:32,341 --> 00:09:33,981
And not hit the button that says launch

194
00:09:33,981 --> 00:09:35,221
Oh

195
00:09:35,221 --> 00:09:36,941
Alright

196
00:09:43,541 --> 00:09:45,541
我认为了一些语言

197
00:09:45,541 --> 00:09:48,541
我认为了一些语言

198
00:09:48,541 --> 00:09:49,541
我认为了一些语言

199
00:09:49,541 --> 00:09:50,541
我认为了语言

200
00:09:50,541 --> 00:09:51,541
语言语言

201
00:09:51,541 --> 00:09:52,541
语言语言

202
00:09:52,541 --> 00:09:53,541
语言语言

203
00:09:53,541 --> 00:09:54,541
语言语言

204
00:09:54,541 --> 00:09:55,541
语言语言

205
00:09:55,541 --> 00:09:56,541
语言语言

206
00:09:56,541 --> 00:09:57,541
语言语言

207
00:09:57,541 --> 00:09:58,541
语言语言

208
00:09:58,541 --> 00:09:59,541
语言语言

209
00:09:59,541 --> 00:10:00,541
语言语言

210
00:10:00,541 --> 00:10:01,541
语言语言

211
00:10:01,541 --> 00:10:02,541
语言语言

212
00:10:02,541 --> 00:10:03,541
语言语言

213
00:10:03,541 --> 00:10:04,541
语言语言

214
00:10:04,541 --> 00:10:05,541
语言语言

215
00:10:05,541 --> 00:10:06,541
语言语言

216
00:10:06,541 --> 00:10:07,541
语言语言

217
00:10:07,541 --> 00:10:08,541
语言语言

218
00:10:08,541 --> 00:10:09,541
语言语言

219
00:10:09,541 --> 00:10:10,541
语言语言

220
00:10:10,541 --> 00:10:11,541
语言语言

221
00:10:11,541 --> 00:10:12,541
语言语言语言

222
00:10:12,541 --> 00:10:16,541
but I applaud it for its intricacy

223
00:10:16,541 --> 00:10:17,821
and then oh read me

224
00:10:17,821 --> 00:10:20,861
okay so our special feature is something like holographic

225
00:10:20,861 --> 00:10:23,181
and I would imagine having a couple apps open

226
00:10:23,181 --> 00:10:25,481
just when we trigger that is probably the move

227
00:10:25,481 --> 00:10:28,281
so oh yeah all right

228
00:10:28,281 --> 00:10:30,301
so it just puts like an interesting card effect

229
00:10:30,301 --> 00:10:31,321
this is what it looks like

230
00:10:31,321 --> 00:10:34,021
this looks like an Apple Vision Pro like app demo

231
00:10:34,021 --> 00:10:37,161
this is the way they look from inside the Vision Pro headset

232
00:10:37,161 --> 00:10:40,361
just if you have used one

233
00:10:40,361 --> 00:10:42,281
你也有意味着,

234
00:10:42,281 --> 00:10:42,881
你也有意味着,

235
00:10:42,881 --> 00:10:44,201
OK,

236
00:10:44,201 --> 00:10:44,321
OK,

237
00:10:44,321 --> 00:10:44,881
OK,

238
00:10:44,881 --> 00:10:45,281
OK,

239
00:10:45,281 --> 00:10:45,401
OK,

240
00:10:45,401 --> 00:10:45,481
OK,

241
00:10:45,481 --> 00:10:45,921
OK,

242
00:10:45,921 --> 00:10:46,361
OK,

243
00:10:46,361 --> 00:10:46,841
OK,

244
00:10:46,841 --> 00:10:46,921
OK,

245
00:10:46,921 --> 00:10:47,401
OK,

246
00:10:47,401 --> 00:10:47,481
OK,

247
00:10:47,481 --> 00:10:48,481
OK,

248
00:10:48,481 --> 00:11:18,481
但那是一渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐渐�

249
00:11:18,481 --> 00:11:20,481
一段时间是在继续上的一段时间

250
00:11:20,481 --> 00:11:23,921
所以,在68秒24秒

251
00:11:23,921 --> 00:11:25,921
SKATENYC是完成的

252
00:11:25,921 --> 00:11:28,081
所以,你连你了解写的

253
00:11:28,081 --> 00:11:29,921
你连你了解写的

254
00:11:29,921 --> 00:11:31,921
你还没有看写的

255
00:11:33,761 --> 00:11:34,881
我感到有点兴趣

256
00:11:34,881 --> 00:11:36,481
现在我们知道

257
00:11:36,481 --> 00:11:38,441
我现在已经有点点看

258
00:11:38,441 --> 00:11:40,481
我说了什么比较多

259
00:11:40,481 --> 00:11:42,081
我们就看一看

260
00:11:42,081 --> 00:11:44,081
在这一段时间

261
00:11:44,081 --> 00:11:45,681
所以,我们来到

262
00:11:45,681 --> 00:11:46,981
NYC Skate Game

263
00:11:46,981 --> 00:11:48,581
should be in outputs, right?

264
00:11:48,701 --> 00:11:48,921
yes

265
00:11:48,921 --> 00:11:51,701
alright

266
00:11:51,701 --> 00:11:54,401
it's not quite

267
00:11:54,401 --> 00:11:56,121
as good as the

268
00:11:56,121 --> 00:11:57,481
Quen 3.8 model

269
00:11:57,481 --> 00:11:57,981
but

270
00:11:57,981 --> 00:12:00,661
that model does have

271
00:12:00,661 --> 00:12:01,881
800 billion more parameters

272
00:12:01,881 --> 00:12:05,461
does that say

273
00:12:05,461 --> 00:12:07,081
NYC Skate on the ground?

274
00:12:07,221 --> 00:12:07,801
it's inverted

275
00:12:07,801 --> 00:12:08,481
but it does

276
00:12:08,481 --> 00:12:08,881
and it says

277
00:12:08,881 --> 00:12:10,181
established 1999

278
00:12:10,181 --> 00:12:11,021
alright

279
00:12:11,021 --> 00:12:11,561
uh oh

280
00:12:11,561 --> 00:12:13,361
sounded like the beginning

281
00:12:13,361 --> 00:12:13,981
of uh

282
00:12:13,981 --> 00:12:15,981
the Van Halen song

283
00:12:15,981 --> 00:12:18,141
I don't know what exactly is

284
00:12:18,141 --> 00:12:19,061
Alright

285
00:12:19,061 --> 00:12:21,561
let's just maybe not

286
00:12:21,561 --> 00:12:22,781
do that same thing

287
00:12:22,781 --> 00:12:24,321
there are sound effects though

288
00:12:24,321 --> 00:12:25,741
I don't normally hear sound effects

289
00:12:25,741 --> 00:12:27,021
can we actually move beyond

290
00:12:27,021 --> 00:12:29,101
Alright not quite

291
00:12:29,101 --> 00:12:30,421
I will say

292
00:12:30,421 --> 00:12:32,321
it's not wonderful

293
00:12:32,321 --> 00:12:34,161
that's actually kind of cool though

294
00:12:34,161 --> 00:12:35,101
the way it like

295
00:12:35,101 --> 00:12:37,581
stylized the game title

296
00:12:37,581 --> 00:12:38,721
Nice

297
00:12:38,721 --> 00:12:39,361
Nice

298
00:12:39,361 --> 00:12:43,941
Well we don't need to do the drum kit sim

299
00:12:43,941 --> 00:12:46,641
the trees are

300
00:12:46,641 --> 00:12:51,041
quite reminiscent of a poo

301
00:12:51,041 --> 00:12:53,281
if i were to be honest

302
00:12:53,281 --> 00:12:55,801
but alright the legs are like kicking

303
00:12:55,801 --> 00:13:04,661
this may be the funniest skate game result we've ever seen

304
00:13:04,661 --> 00:13:06,201
which is you know something

305
00:13:06,201 --> 00:13:07,821
we have a street sign there

306
00:13:07,821 --> 00:13:10,881
the text is a little sideways

307
00:13:10,881 --> 00:13:12,381
as are the building faces

308
00:13:12,381 --> 00:13:16,661
this is definitely a troubled result however i do like the water fountain

309
00:13:16,661 --> 00:13:19,201
shooting out i believe that was something i've seen

310
00:13:19,201 --> 00:13:28,761
i've seen that in i think the quen result now here's what i would like to just quickly do for

311
00:13:28,761 --> 00:13:34,841
some reference this is the result we received for the c++ skateboard test from the preview version

312
00:13:34,841 --> 00:13:41,661
of deep seek v4 we can see right here i'm going to turn the speaker off this is pretty bad it was

313
00:13:41,661 --> 00:13:48,081
like so bad that I actually remember to go to this specific thing just to showcase it so the leap in

314
00:13:48,081 --> 00:13:54,041
capability based off of what we see right here to what we just saw in the skate game that it had

315
00:13:54,041 --> 00:13:59,541
created for us which we can just look at in these screenshots here is still pretty significant next

316
00:13:59,541 --> 00:14:04,081
up I'm giving it the subway fps test I've also added on to the end of the prompt don't worry

317
00:14:04,081 --> 00:14:08,161
about checking over the script I will manually check it and if there are issues I will report

318
00:14:08,161 --> 00:14:13,541
them back to you just because this will really like extend every single result by like two times

319
00:14:13,541 --> 00:14:19,381
in terms of like time wise length just in the interest of something I could do visually in

320
00:14:19,381 --> 00:14:23,601
like five seconds and then be like hey this doesn't work so we're gonna do that just to maybe like

321
00:14:23,601 --> 00:14:29,261
keep things a little tighter all right so it did finish the subway fps game the problem is it just

322
00:14:29,261 --> 00:14:34,441
keeps trying to do like troubleshooting which it's not good at doing on this system for some

323
00:14:34,441 --> 00:14:41,561
reason all right let's check out our subway fps mont rand avenue all right i'm gonna turn the

324
00:14:41,561 --> 00:14:49,301
speaker on so it gave us two options we can either explore the station okay it's hey you know what

325
00:14:49,301 --> 00:14:57,281
i think that is the best subway car that i've ever seen drawn in this test and you're gonna be like

326
00:14:57,281 --> 00:15:01,961
well it doesn't look that great i know it's just that they never really include these and when they

327
00:15:01,961 --> 00:15:03,961
如果那些人都不太好,

328
00:15:03,961 --> 00:15:05,961
所以这其实...

329
00:15:05,961 --> 00:15:07,961
如果那些人开始移动,

330
00:15:07,961 --> 00:15:10,961
我会会成为更多感激,

331
00:15:10,961 --> 00:15:12,961
而是很简单的,

332
00:15:12,961 --> 00:15:16,961
这些人是非常清晰的,

333
00:15:16,961 --> 00:15:18,961
它是非常简单的,

334
00:15:18,961 --> 00:15:19,961
它是完全不断坏的,

335
00:15:19,961 --> 00:15:21,961
我会有横移动的,

336
00:15:21,961 --> 00:15:22,961
我们有横移动,

337
00:15:22,961 --> 00:15:24,961
我们有横移动,

338
00:15:24,961 --> 00:15:25,961
我们有横移动,

339
00:15:25,961 --> 00:15:27,961
我们有横移动,

340
00:15:27,961 --> 00:15:28,961
然后,

341
00:15:28,961 --> 00:15:33,921
导致,像是指示的,像是水平的,

342
00:15:33,921 --> 00:15:36,561
水平的水平,更簡單,

343
00:15:36,561 --> 00:15:39,561
而且我們不能往上的水平,我們不想要去

344
00:15:39,561 --> 00:15:44,761
現在,當然,另一個其他東西是FPS的,

345
00:15:44,761 --> 00:15:48,561
所以,我們回到 menu,然後入了FPS,

346
00:15:48,561 --> 00:15:50,361
我們看看能否改變,

347
00:15:50,361 --> 00:15:52,761
噢,我的,那是一個好玩的,

348
00:15:52,761 --> 00:15:56,161
它是,它是,我看,

349
00:15:56,161 --> 00:15:58,161
我看了

350
00:15:58,161 --> 00:16:00,161
我看了

351
00:16:00,161 --> 00:16:02,161
它是很滑的

352
00:16:02,161 --> 00:16:04,161
我看不見任何人

353
00:16:04,161 --> 00:16:06,161
看不見

354
00:16:06,161 --> 00:16:08,161
我看不見

355
00:16:08,161 --> 00:16:10,161
所以

356
00:16:10,161 --> 00:16:12,161
我看不見

357
00:16:12,161 --> 00:16:14,161
有什麼人的

358
00:16:14,303 --> 00:16:29,303
對,它是滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴滴�

359
00:16:29,303 --> 00:16:30,303
它是在下方的下方,

360
00:16:30,303 --> 00:16:32,303
所以我把它给它一段时间。

361
00:16:32,303 --> 00:16:34,303
我停止它,

362
00:16:34,303 --> 00:16:36,303
因为它已经停止了,

363
00:16:36,303 --> 00:16:37,303
它是在这种程度上,

364
00:16:37,303 --> 00:16:39,303
它可以用它在下方的游戏,

365
00:16:39,303 --> 00:16:40,303
所以在这种程度,

366
00:16:40,303 --> 00:16:41,303
它是更容易的,

367
00:16:41,303 --> 00:16:44,303
所以我给它那些评论。

368
00:16:44,303 --> 00:16:45,303
所以,我们的评论,

369
00:16:45,303 --> 00:16:46,303
我们的设计,

370
00:16:46,303 --> 00:16:47,303
我们的设计,

371
00:16:47,303 --> 00:16:49,303
是完成了。

372
00:16:49,303 --> 00:16:50,303
我们就看一看,

373
00:16:50,303 --> 00:16:52,303
它是在。

374
00:16:52,303 --> 00:16:53,303
好,

375
00:16:53,303 --> 00:16:54,303
我看它是在。

376
00:16:54,303 --> 00:16:56,303
我看它是在的位置,

377
00:16:56,303 --> 00:16:57,303
所以,

378
00:16:57,303 --> 00:16:58,303
好,

379
00:16:58,303 --> 00:17:19,783
好了,七 hostels, it's showing me in the top right, I unfortunately just can't locate them, I would have expected it to have done better, and again this is one where we could go back and reference the previous model of this, so basically the preview version, it was significantly less detailed than this was.

380
00:17:19,783 --> 00:17:27,583
though unfortunately we don't have enemies and given the amount of time it took as well as the

381
00:17:27,583 --> 00:17:32,183
kind of multiple times i told that there were no enemies i'm regrettably going to call this one

382
00:17:32,183 --> 00:17:38,583
next up i'm giving this a difficult task to create a 3d model of an rb26 inline 6 engine

383
00:17:38,583 --> 00:17:44,183
it must be something that can be 3d printed with little to no support and fit inside of it a tiny

384
00:17:44,183 --> 00:17:49,263
dc motor called an n20 motor because this is something that i've told it should like work

385
00:17:49,263 --> 00:17:50,543
with my RC car project.

386
00:17:50,863 --> 00:17:52,923
Actually, I don't think that's specifically in this prompt,

387
00:17:53,023 --> 00:17:54,203
but that's the whole point of this,

388
00:17:54,503 --> 00:17:56,483
where this is, it should understand

389
00:17:56,483 --> 00:17:58,323
and it will scale it properly to that motor.

390
00:17:58,723 --> 00:18:01,783
So far, I've run this test with a few intelligent models

391
00:18:01,783 --> 00:18:03,323
and I've never really received anything

392
00:18:03,323 --> 00:18:04,483
that was actually worth printing.

393
00:18:04,623 --> 00:18:07,483
So hopefully sooner than later, that will change.

394
00:18:07,643 --> 00:18:10,003
So our 3D model task is pretty interesting

395
00:18:10,003 --> 00:18:11,823
because it's just doing a lot of research

396
00:18:11,823 --> 00:18:13,983
on like the actual proper orientation

397
00:18:13,983 --> 00:18:15,983
of specific items on this engine.

398
00:18:16,183 --> 00:18:18,863
Like where's the downpipe, which is an exhaust component.

399
00:18:18,863 --> 00:18:23,823
what side is it what is the intake side what is the exhaust side intake plenum left or right so

400
00:18:23,823 --> 00:18:29,483
i do like to see the amount of like intricate research that's going into the culmination of

401
00:18:29,483 --> 00:18:35,743
this model creation and it would be really awesome for like us to get a model for this task that's

402
00:18:35,743 --> 00:18:41,443
actually printable it would be the first one so in one hour and 10 minutes we received our 3d model

403
00:18:41,443 --> 00:18:47,883
response this is a very very intricate just bit of text about the thing it made now of course the

404
00:18:47,883 --> 00:18:50,883
另一部分是在看一看,

405
00:18:50,883 --> 00:18:53,883
但我記得我把這個名字的成功,

406
00:18:53,883 --> 00:18:54,883
我們會可以看一看,

407
00:18:54,883 --> 00:18:55,883
整個設計設計STL,

408
00:18:55,883 --> 00:18:57,883
我會想到這會的情況,

409
00:18:57,883 --> 00:18:59,883
請你來個好,

410
00:18:59,883 --> 00:19:01,883
我需要一個好好的一件,

411
00:19:01,883 --> 00:19:04,883
其實,

412
00:19:04,883 --> 00:19:06,883
它是相似的,

413
00:19:06,883 --> 00:19:07,883
這真的很好,

414
00:19:07,883 --> 00:19:09,883
這真的很好,

415
00:19:09,883 --> 00:19:11,883
這真的很好,

416
00:19:11,883 --> 00:19:14,883
這真的很好,

417
00:19:14,883 --> 00:19:23,123
it's not perfect we have some sideways turbocharger issues though i mean this is all right let's let's

418
00:19:23,123 --> 00:19:27,603
peek at this in open scad there are a bunch of individual pieces for this so we also want to see

419
00:19:27,603 --> 00:19:32,423
if the engine that we told it needs to fit inside the little remote control car engine let's see if

420
00:19:32,423 --> 00:19:38,663
it has any hope of properly fitting research it has a research folder right here just showing like

421
00:19:38,663 --> 00:19:41,663
很多的照片和资料

422
00:19:41,663 --> 00:19:43,663
非常好

423
00:19:43,663 --> 00:19:46,663
而且有些更多的预线

424
00:19:46,663 --> 00:19:49,663
这一定是非常好做

425
00:19:49,663 --> 00:19:52,663
Turbo side

426
00:19:52,663 --> 00:19:53,663
underside

427
00:19:53,663 --> 00:19:56,663
不算是很容易

428
00:19:56,663 --> 00:19:58,663
可以3D刷子

429
00:19:58,663 --> 00:20:00,663
我认为我可以做这个

430
00:20:00,663 --> 00:20:02,663
因为这些组织

431
00:20:02,663 --> 00:20:04,663
所以它不是在一样

432
00:20:04,663 --> 00:20:06,663
就在一样一样

433
00:20:06,663 --> 00:20:36,663
一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一共一

434
00:20:36,663 --> 00:20:40,223
我认为这些是有关的帖子,这些是很酷的

435
00:20:40,223 --> 00:20:43,903
这些是可以带到这些帖子,没有任何支援的帖子

436
00:20:43,903 --> 00:20:45,903
如果我们把它放开

437
00:20:46,543 --> 00:20:51,263
这些帖子也有在买的字符,或者是在买的字符,买上的字符

438
00:20:53,303 --> 00:20:56,663
这些是很好的,没有问题

439
00:20:56,663 --> 00:20:59,663
油 pen,这些是我们的

440
00:20:59,663 --> 00:21:02,663
这些是我们的帖子,我们不准备去带这个

441
00:21:02,663 --> 00:21:10,223
I'm not disappointed but I'm not super impressed either so I suppose I am lukewarm when it comes

442
00:21:10,223 --> 00:21:15,763
to judging this definitely if we just look at this as a render even forget about looking at

443
00:21:15,763 --> 00:21:20,963
the individual parts there because it's not really going to be feasible to print this it did a really

444
00:21:20,963 --> 00:21:25,783
decent job just assembling it the research that went into this I think is almost probably like

445
00:21:25,783 --> 00:21:30,743
I hate to sound like an AI here but like the story isn't the model it's the research that

446
00:21:30,743 --> 00:21:37,463
went into it that's like generic like ai slop it really did a lot of decent research here i mean

447
00:21:37,463 --> 00:21:43,823
did it find this and download it just to see no okay so this is something it maybe made or maybe

448
00:21:43,823 --> 00:21:49,323
probably found online just to be able to better assess what the size of this asset is but it took

449
00:21:49,323 --> 00:21:53,083
a lot of care to ensure that what it tried to generate would have hypothetically fit

450
00:21:53,083 --> 00:22:00,403
the specific specifications that were denoted in the original prompt so not half bad i'm also going

451
00:22:00,403 --> 00:22:05,063
to give this a simple frontend design test where it needs to create the watch website so it has to

452
00:22:05,063 --> 00:22:10,523
have a pretty cinematic panning effect over a 3d watch asset that it creates i didn't specifically

453
00:22:10,523 --> 00:22:15,423
okay no i did specifically denote that the watch needs to be a 3d model so we'll see how competent

454
00:22:15,423 --> 00:22:21,203
it is at putting the entirety of this together so in one hour and 30 minutes we got our watch

455
00:22:21,203 --> 00:22:26,523
frontend result it again seems to have a pretty substantial like just list of what was built here

456
00:22:26,523 --> 00:22:28,443
so let's take a look and see how it is

457
00:22:28,443 --> 00:22:30,503
beautiful watch front end website

458
00:22:30,503 --> 00:22:33,403
that's actually

459
00:22:33,403 --> 00:22:35,643
so close to excellent

460
00:22:35,643 --> 00:22:37,723
the big problem is this

461
00:22:37,723 --> 00:22:40,403
I would assume is the watch strap

462
00:22:40,403 --> 00:22:41,003
sticking out

463
00:22:41,003 --> 00:22:44,523
this is 3D so don't be fooled by its 2Dishness

464
00:22:44,523 --> 00:22:44,803
look

465
00:22:44,803 --> 00:22:46,503
it is actually 3D

466
00:22:46,503 --> 00:22:47,923
it's slowly panning too

467
00:22:47,923 --> 00:22:50,643
okay so a couple things I do like

468
00:22:50,643 --> 00:22:51,963
the wooden background

469
00:22:51,963 --> 00:22:54,603
and then this leather card or pad

470
00:22:54,603 --> 00:22:56,043
that this watch is placed on

471
00:22:56,043 --> 00:22:58,043
另外的那,看起來很像很像,

472
00:22:58,043 --> 00:23:00,043
看起來很像很像,

473
00:23:00,043 --> 00:23:02,043
看起來很像,

474
00:23:02,043 --> 00:23:04,043
它是一種高的衣服,

475
00:23:04,043 --> 00:23:07,043
是正在我們做的一種試試,

476
00:23:07,043 --> 00:23:08,043
現在,

477
00:23:08,043 --> 00:23:09,043
那其實,

478
00:23:09,043 --> 00:23:11,043
那裡看起來很像,

479
00:23:11,043 --> 00:23:12,043
看起來很像,

480
00:23:12,043 --> 00:23:14,043
看起來很像,

481
00:23:14,043 --> 00:23:16,043
看起來很像,

482
00:23:16,043 --> 00:23:18,043
看起來很像,

483
00:23:18,043 --> 00:23:20,043
那裡也應該不像,

484
00:23:20,043 --> 00:23:21,043
但,

485
00:23:21,043 --> 00:23:23,043
看起來很像,

486
00:23:23,043 --> 00:23:25,043
看起來很像,

487
00:23:25,043 --> 00:23:33,043
we start to scroll down quiet luxury measured in seconds sunburst dial okay it has like a material

488
00:23:33,043 --> 00:23:38,143
card here the sapphire crystal is perhaps something i would have maybe omitted from this section

489
00:23:38,143 --> 00:23:43,903
but the dial looks good two watches one point of view please render them oh look at that that is

490
00:23:43,903 --> 00:23:48,603
sick that really looks pretty good i love the way they're placed on the table really the big thing

491
00:23:48,603 --> 00:23:53,403
letting it down is just this leather slab is inhibiting our view of these beautiful straps

492
00:23:53,403 --> 00:23:59,443
this one is the stainless steel option this one is the gold option and priced accordingly i would

493
00:23:59,443 --> 00:24:06,363
imagine yeah although i would say the jump in price of like that is the negligible amount to

494
00:24:06,363 --> 00:24:10,643
get the gold one i'd probably go with this one personally oh cool and these ones we can freely

495
00:24:10,643 --> 00:24:11,923
rotate so that helps a bit

496
00:24:27,443 --> 00:24:29,443
它們的一種類型的質素

497
00:24:29,443 --> 00:24:31,443
所以它們不是一種類型的質素

498
00:24:31,443 --> 00:24:32,883
它們是銀色的

499
00:24:32,883 --> 00:24:34,443
然後銀色的

500
00:24:34,443 --> 00:24:35,443
然後銀色的

501
00:24:35,443 --> 00:24:36,443
然後銀色的

502
00:24:36,443 --> 00:24:37,443
所以那是一些注意力

503
00:24:37,443 --> 00:24:38,443
我喜歡看的

504
00:24:38,443 --> 00:24:39,443
更加上的

505
00:24:39,443 --> 00:24:41,443
一種類型的質素

506
00:24:41,443 --> 00:24:43,443
是在這裏面上的

507
00:24:43,443 --> 00:24:46,443
這裏面上的3D看

508
00:24:46,443 --> 00:24:47,443
但它們的顯示

509
00:24:47,443 --> 00:24:49,443
它們的效果

510
00:24:49,443 --> 00:24:50,443
那看起來很不錯

511
00:24:50,443 --> 00:24:51,443
然後我們有一個

512
00:24:51,443 --> 00:24:52,443
綠色的

513
00:24:52,443 --> 00:24:53,443
我喜歡這個

514
00:24:53,443 --> 00:24:54,443
我喜歡這個

515
00:24:54,443 --> 00:24:55,443
甚至是

516
00:24:55,443 --> 00:24:56,443
不完美

517
00:24:56,443 --> 00:24:58,443
就是完美了

518
00:25:26,443 --> 00:25:28,443
我可以做一些事情

519
00:25:28,443 --> 00:25:30,443
我們可以做一些事情

520
00:25:30,443 --> 00:25:32,443
我們要做 Steve the PC Repairman

521
00:25:32,443 --> 00:25:34,443
遊戲中的遊戲

522
00:25:34,443 --> 00:25:36,443
大概65分鐘

523
00:25:36,443 --> 00:25:38,443
所以在一小時

524
00:25:38,443 --> 00:25:40,443
我會做什麼

525
00:25:40,443 --> 00:25:42,443
我會做什麼

526
00:25:42,443 --> 00:25:44,443
不想它是在這裏

527
00:25:44,443 --> 00:25:46,443
所以

528
00:25:46,443 --> 00:25:47,443
1

529
00:25:47,443 --> 00:25:49,443
Azure Palm Job

530
00:25:49,443 --> 00:25:50,443
OK

531
00:25:50,443 --> 00:25:51,443
Click to begin

532
00:25:51,443 --> 00:25:52,443
開動

533
00:25:52,443 --> 00:25:54,443
我們有一個 speaker

534
00:25:54,443 --> 00:25:55,443
好

535
00:25:59,443 --> 00:26:00,443
這就是好

536
00:26:01,443 --> 00:26:02,443
好,OK

537
00:26:02,443 --> 00:26:03,443
那...what's up with the...

538
00:26:03,443 --> 00:26:04,443
那...what's up with the...

539
00:26:11,443 --> 00:26:12,443
那是正事

540
00:26:12,443 --> 00:26:15,643
oh thank you dear

541
00:26:15,643 --> 00:26:18,343
the screen kept telling me the date was 1980

542
00:26:18,343 --> 00:26:19,723
and i thought

543
00:26:19,723 --> 00:26:20,743
all fixed

544
00:26:20,743 --> 00:26:22,663
i have been time traveling long enough

545
00:26:22,663 --> 00:26:23,183
fresh battery

546
00:26:23,183 --> 00:26:25,243
your computer now believes

547
00:26:25,243 --> 00:26:26,923
i will bring you oatmeal cookies on tuesday

548
00:26:26,923 --> 00:26:28,143
same as the rest of us

549
00:26:28,143 --> 00:26:29,923
and give mr buttons a scratch behind the ears for me

550
00:26:29,923 --> 00:26:32,643
let me walk you out ma'am

551
00:26:32,643 --> 00:26:35,323
i'll be right with you

552
00:26:35,323 --> 00:26:36,523
mr thomas

553
00:26:36,523 --> 00:26:38,703
the speech overlays itself

554
00:26:38,703 --> 00:26:39,943
you are a good boy steven

555
00:26:39,943 --> 00:26:40,703
oh

556
00:26:40,703 --> 00:26:43,903
如果那些人不需要的,你会有空的空间

557
00:26:43,903 --> 00:26:44,983
你会有空间的空间

558
00:26:44,983 --> 00:26:46,223
我喜欢空间的空间

559
00:26:46,223 --> 00:26:47,423
我会有空间的空间

560
00:26:47,423 --> 00:26:49,463
我会有空间的空间

561
00:26:49,463 --> 00:26:51,023
我会有空间的空间

562
00:26:51,023 --> 00:26:52,943
我会有空间的空间

563
00:26:52,943 --> 00:26:54,783
我会有空间的空间

564
00:26:54,783 --> 00:26:57,023
你看到他在空间的空间的空间

565
00:26:57,023 --> 00:26:58,383
他会有空间的空间

566
00:26:58,383 --> 00:26:59,663
有空间的空间

567
00:26:59,663 --> 00:27:01,103
我会有空间的空间

568
00:27:01,103 --> 00:27:02,303
我会有空间的空间

569
00:27:02,303 --> 00:27:03,223
他会有空间的空间

570
00:27:03,223 --> 00:27:05,023
他会有空间的空间

571
00:27:05,023 --> 00:27:06,943
他会有空间的空间

572
00:27:06,943 --> 00:27:08,983
他会有空间的空间

573
00:27:08,983 --> 00:27:38,983
我会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想到你会想

574
00:27:38,983 --> 00:27:40,823
剪剪 over

575
00:27:40,823 --> 00:27:42,083
thank goodness

576
00:27:42,083 --> 00:27:43,363
click to take control

577
00:27:43,363 --> 00:27:48,243
aside from the sounds being just totally messed up

578
00:27:48,243 --> 00:27:49,203
this is actually

579
00:27:49,203 --> 00:27:52,223
pretty decent

580
00:27:52,223 --> 00:27:53,523
okay

581
00:27:53,523 --> 00:27:54,643
the cat is

582
00:27:54,643 --> 00:27:58,463
there's definitely some

583
00:27:58,463 --> 00:28:00,743
geometric issues

584
00:28:00,743 --> 00:28:02,243
what are we supposed to do

585
00:28:02,243 --> 00:28:04,223
take the briefcase when you're ready

586
00:28:04,223 --> 00:28:05,603
okay where is it

587
00:28:05,603 --> 00:28:07,863
we'll pet buttons

588
00:28:07,863 --> 00:28:11,923
don't look at me like that buttons

589
00:28:11,923 --> 00:28:12,803
easter egg

590
00:28:12,803 --> 00:28:13,943
man's gotta eat

591
00:28:13,943 --> 00:28:16,463
okay where's the briefcase

592
00:28:16,463 --> 00:28:18,243
all right

593
00:28:18,243 --> 00:28:21,083
we have inventory zip ties

594
00:28:21,083 --> 00:28:21,663
okay

595
00:28:21,663 --> 00:28:23,263
e to continue

596
00:28:23,263 --> 00:28:24,623
all right is it gonna put us

597
00:28:24,623 --> 00:28:24,923
Miami

598
00:28:24,923 --> 00:28:25,523
nice

599
00:28:25,523 --> 00:28:26,363
palm trees

600
00:28:26,363 --> 00:28:27,103
humidity

601
00:28:27,103 --> 00:28:29,223
and one very bad man

602
00:28:29,223 --> 00:28:30,863
who preys on people who knit

603
00:28:30,863 --> 00:28:31,983
champagne

604
00:28:31,983 --> 00:28:32,483
perhaps

605
00:28:32,483 --> 00:28:33,383
I'm just gonna click through

606
00:28:33,383 --> 00:28:33,983
because it's like

607
00:28:33,983 --> 00:28:35,063
black

608
00:28:35,063 --> 00:28:36,203
like my

609
00:28:36,203 --> 00:28:36,863
well

610
00:28:36,863 --> 00:28:37,843
coffee

611
00:28:37,843 --> 00:29:07,823
赤字幕 李宗盛

612
00:29:07,843 --> 00:29:08,843
一枚的椅子

613
00:29:08,843 --> 00:29:09,843
哇

614
00:29:09,843 --> 00:29:11,643
這其實是很像的

615
00:29:11,643 --> 00:29:12,003
intricate

616
00:29:12,003 --> 00:29:12,483
好, good

617
00:29:12,483 --> 00:29:14,003
我怕我們不可以

618
00:29:14,003 --> 00:29:15,483
走過的壁

619
00:29:15,483 --> 00:29:15,843
但

620
00:29:15,843 --> 00:29:17,003
我們可以

621
00:29:17,003 --> 00:29:17,803
把蟲子放在

622
00:29:17,803 --> 00:29:18,803
你的風格

623
00:29:18,803 --> 00:29:19,403
沒有什麼

624
00:29:19,403 --> 00:29:20,203
CMOS errors

625
00:29:20,203 --> 00:29:20,883
5050

626
00:29:20,883 --> 00:29:22,443
這看起來

627
00:29:22,443 --> 00:29:23,363
Steve's shop

628
00:29:23,363 --> 00:29:23,843
在船上的

629
00:29:23,843 --> 00:29:24,803
船上的船上

630
00:29:24,803 --> 00:29:25,203
是在

631
00:29:25,203 --> 00:29:27,523
遇到的地方

632
00:29:27,523 --> 00:29:29,043
我們應該要

633
00:29:29,043 --> 00:29:29,883
我們就

634
00:29:29,883 --> 00:29:30,483
ignore that

635
00:29:30,483 --> 00:29:31,923
所以

636
00:29:31,923 --> 00:29:32,803
我們在

637
00:29:32,803 --> 00:29:34,083
前面

638
00:29:34,083 --> 00:29:34,283
然後

639
00:29:34,283 --> 00:29:35,483
到17層

640
00:29:35,483 --> 00:30:05,463
我们现在有个屏幕

641
00:30:05,483 --> 00:30:09,983
what is that

642
00:30:09,983 --> 00:30:14,303
reach okay west service stairwell

643
00:30:14,303 --> 00:30:16,923
let's just assume this is north

644
00:30:16,923 --> 00:30:21,023
good

645
00:30:35,483 --> 00:31:05,463
我叫你

646
00:31:05,483 --> 00:31:35,483
我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说,我们来说

647
00:31:35,483 --> 00:31:37,483
是,當然,這遊戲的一種類似的,

648
00:31:37,483 --> 00:31:40,483
這就是這張文字的文字,

649
00:31:40,483 --> 00:31:41,483
在這張文字上,

650
00:31:41,483 --> 00:31:42,483
這張文字上,

651
00:31:42,483 --> 00:31:43,483
完全不平的文字,

652
00:31:43,483 --> 00:31:45,483
令我感到不平的文字,

653
00:31:45,483 --> 00:31:47,483
所以,這張是個Bijan的問題

654
00:31:47,991 --> 00:31:50,991
好,好,我忘了…

655
00:31:50,991 --> 00:31:54,651
I'm just running in a server one fist zero mercy

656
00:31:54,651 --> 00:31:59,631
I already like this oh it seems like

657
00:31:59,631 --> 00:32:06,631
good music all right all right all right

658
00:32:06,631 --> 00:32:12,331
let's just explore the city real quick

659
00:32:12,331 --> 00:32:20,371
I don't even know okay so I saw a car get like

660
00:32:20,371 --> 00:32:26,971
这一种是真的有趣的玩意

661
00:32:26,971 --> 00:32:35,471
excuse me

662
00:32:35,471 --> 00:32:36,751
excuse me

663
00:32:36,751 --> 00:32:37,851
pardon me

664
00:32:37,851 --> 00:32:38,971
it does like a

665
00:32:38,971 --> 00:32:39,631
like a

666
00:32:39,631 --> 00:32:41,171
okay

667
00:32:41,171 --> 00:32:42,391
i don't know that we hit

668
00:32:42,391 --> 00:32:42,991
is that a

669
00:32:42,991 --> 00:32:44,011
can we hit the bike

670
00:32:44,011 --> 00:32:48,131
bonk

671
00:32:48,131 --> 00:32:54,971
呃呦

672
00:32:54,971 --> 00:32:56,391
Jaywalking Pedestrian

673
00:32:56,391 --> 00:33:02,391
can we punch this parking meter

674
00:33:02,391 --> 00:33:03,351
yeah we can just

675
00:33:03,351 --> 00:33:04,951
everything except the buildings

676
00:33:04,951 --> 00:33:05,891
alright there's

677
00:33:05,891 --> 00:33:07,531
unfortunately the orientation

678
00:33:07,531 --> 00:33:08,671
of some of these things

679
00:33:08,671 --> 00:33:09,531
is not quite right

680
00:33:09,531 --> 00:33:11,411
we can see that we're moving sideways

681
00:33:11,411 --> 00:33:12,511
when we move forward

682
00:33:12,511 --> 00:33:17,191
and the cars are also

683
00:33:17,191 --> 00:33:19,871
sorry pal

684
00:33:19,871 --> 00:33:24,391
and it's interesting how they

685
00:33:24,391 --> 00:33:26,591
interact with the scenery as well

686
00:33:26,591 --> 00:33:27,871
can we eat this statue

687
00:33:27,871 --> 00:33:29,131
I don't think we can but

688
00:33:29,131 --> 00:33:30,731
oh yeah we can

689
00:33:30,731 --> 00:33:36,611
I will say the cartoonness of this game

690
00:33:36,611 --> 00:33:38,251
is actually kind of well done

691
00:33:38,251 --> 00:33:40,431
I don't know how easy that is to assess

692
00:33:40,431 --> 00:33:41,831
just watching someone play it

693
00:33:41,831 --> 00:33:43,931
but like the pacing is decent

694
00:33:43,931 --> 00:33:46,831
I think maybe that's attributed to the music as well

695
00:33:46,831 --> 00:33:52,091
but the way that when you actually are the loop is getting a little repetitive when you actually

696
00:33:52,091 --> 00:33:57,171
eat the things and they also go and interact with other items in the environment so like how

697
00:33:57,171 --> 00:34:02,571
that will bounce off the building and then maybe hit other things in the path that's pretty cool

698
00:34:02,571 --> 00:34:11,151
and this was this was actually quite arcade style and playable our scores going up we have combo

699
00:34:11,151 --> 00:34:41,151
超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超過超

700
00:34:41,151 --> 00:34:43,391
especially considering the size of it.

701
00:34:43,711 --> 00:34:45,111
This has a bit more competition,

702
00:34:45,431 --> 00:34:48,131
but again, it is a big leap on its predecessor,

703
00:34:48,371 --> 00:34:49,831
which I guess is what we like to see.

704
00:34:49,951 --> 00:34:52,151
I was impressed a few times with some things,

705
00:34:52,251 --> 00:34:54,151
like basically results within results.

706
00:34:54,291 --> 00:34:56,251
The cinematic Steve the PC repair game,

707
00:34:56,331 --> 00:34:58,971
it actually had some elements to it that were like,

708
00:34:59,031 --> 00:35:00,511
hey, you know, it kind of thought this through.

709
00:35:00,651 --> 00:35:02,191
Just the way when they were speaking,

710
00:35:02,331 --> 00:35:04,171
their heads were like moving back and forth,

711
00:35:04,231 --> 00:35:05,791
not the weird chin thing that was happening.

712
00:35:05,891 --> 00:35:07,311
I'm not sure what happened there.

713
00:35:07,591 --> 00:35:09,771
The music it included in both that game

714
00:35:09,771 --> 00:35:11,891
and then the Street Eat game were actually like,

715
00:35:12,171 --> 00:35:15,131
they definitely fit the scene that it was drawing here.

716
00:35:15,211 --> 00:35:18,191
And it did a good job making the Street Eat game very cartoonish.

717
00:35:18,331 --> 00:35:21,871
And the playability of it felt good, I guess, for lack of a better term.

718
00:35:22,171 --> 00:35:26,771
The model that we got of the engine was actually, it wasn't printable.

719
00:35:26,931 --> 00:35:31,291
So unfortunately, maybe it failed to like check that box in the specific task.

720
00:35:31,571 --> 00:35:33,711
However, if we just look at one of the previews right here,

721
00:35:33,991 --> 00:35:35,591
it did a pretty decent job,

722
00:35:35,711 --> 00:35:38,531
definitely better than the other models I've tested this with.

723
00:35:38,531 --> 00:35:44,211
though I have not tested this specific prompt with any big beefy frontier models yet so keep that in

724
00:35:44,211 --> 00:35:48,431
mind as well but it definitely had some chops here and the amount of research it did to actually

725
00:35:48,431 --> 00:35:53,751
get to that point was quite a lot we have all the things in here downloading these images when it

726
00:35:53,751 --> 00:35:58,411
doesn't have vision so I'm not quite sure what to make of that but nonetheless I think probably the

727
00:35:58,411 --> 00:36:04,791
biggest point of frustration for me is it just goes all out when trying to actually ensure that

728
00:36:04,791 --> 00:36:05,811
每一切都要是正確的

729
00:36:05,811 --> 00:36:07,971
每一终的练习的练习是正确的

730
00:36:07,971 --> 00:36:09,771
和一终的练习的练习

731
00:36:09,771 --> 00:36:11,171
它不需要练习

732
00:36:11,171 --> 00:36:13,411
它不需要练习

733
00:36:13,411 --> 00:36:14,851
而且有一个货帮助

734
00:36:14,851 --> 00:36:17,651
也许你会购习这些练习

735
00:36:17,651 --> 00:36:20,051
它就会是最简单的

736
00:36:20,051 --> 00:36:21,651
要不做得到的效果

737
00:36:21,651 --> 00:36:24,411
不做得到的效果

738
00:36:24,411 --> 00:36:25,651
你看它自己的练习

739
00:36:25,651 --> 00:36:26,671
然后告诉它什么

740
00:36:26,671 --> 00:36:27,551
然后告诉它它要解决

741
00:36:27,551 --> 00:36:29,651
要去做得到的

742
00:36:29,651 --> 00:36:30,571
它们的练习

743
00:36:30,571 --> 00:36:31,811
加了很多时间

744
00:36:31,811 --> 00:36:33,251
和购习的练习

745
00:36:33,251 --> 00:36:39,611
to the task which maybe is more of a consideration when doing zero shot tests like this but I find

746
00:36:39,611 --> 00:36:44,571
that's something I've consistently had frustrations with as well as with the flash model where if I'm

747
00:36:44,571 --> 00:36:49,111
having them make like a street heat it's easier if they just spit the script out I test it and say

748
00:36:49,111 --> 00:36:53,391
hey the buildings are disappeared and then it fixes it rather than it trying to go through

749
00:36:53,391 --> 00:36:57,631
and like well let me try to spin up a headless browser and then I'll take a screenshot and like

750
00:36:57,631 --> 00:37:03,071
do a heat map because it can't actually see it or some other function so it's a little token

751
00:37:03,071 --> 00:37:09,131
I did run it on the highest potential reasoning mode so that could be attributed to it as well but

752
00:37:09,131 --> 00:37:15,031
I still would like to be able to do that without it like going all out to try to make sure it's

753
00:37:15,031 --> 00:37:19,531
polishing every aspect of the script because I want max performance I just don't need that level of

754
00:37:19,531 --> 00:37:25,191
depth so that may be a personal complaint but you know so for the final thing let's take a look at

755
00:37:25,191 --> 00:37:31,151
everything we spent for the entirety of this test we can see right here that all cost $1.91

756
00:37:31,151 --> 00:37:33,151
所以那是挺便宜的

757
00:37:33,151 --> 00:37:36,251
你必须要記住,這就是要提升

758
00:37:36,251 --> 00:37:37,431
他們的API價值

759
00:37:37,431 --> 00:37:38,831
他們說有提升

760
00:37:38,831 --> 00:37:40,751
但我認為

761
00:37:40,751 --> 00:37:41,391
一大小小小的

762
00:37:41,391 --> 00:37:42,551
應該是

763
00:37:42,551 --> 00:37:43,951
這整個整個

764
00:37:43,951 --> 00:37:44,851
這裏是 $8

765
00:37:44,851 --> 00:37:46,511
這裏是還未來的

766
00:37:46,511 --> 00:37:47,991
這裏是不可憐的

767
00:37:47,991 --> 00:37:49,231
這裏是個3.8

768
00:37:49,231 --> 00:37:50,311
max test video

769
00:37:50,311 --> 00:37:51,111
我們做了

770
00:37:51,111 --> 00:37:51,991
這幾個月後

771
00:37:51,991 --> 00:37:52,831
或是一個月的

772
00:37:52,831 --> 00:37:53,271
這裏是

773
00:37:53,271 --> 00:37:54,631
那是$32

774
00:37:54,631 --> 00:37:55,671
所以保留了

775
00:37:55,671 --> 00:37:56,711
這裏是

776
00:37:56,711 --> 00:37:57,271
這裏是

777
00:37:57,271 --> 00:37:57,771
這裏是比較

778
00:37:57,771 --> 00:37:58,351
performant

779
00:37:58,351 --> 00:37:58,991
但這裏是

780
00:37:58,991 --> 00:37:59,831
有依照的

781
00:37:59,831 --> 00:38:04,351
依然是非常非常非常便宜

782
00:38:04,351 --> 00:38:06,031
也不仍然很便宜

783
00:38:06,031 --> 00:38:08,531
但仍然是五次费的费

784
00:38:08,531 --> 00:38:09,631
仍然是不仍然不仍然

785
00:38:09,631 --> 00:38:12,071
我认为这组织的试试

786
00:38:12,071 --> 00:38:15,071
第0813 DeepSeek V4 Pro 这一款版本

787
00:38:15,071 --> 00:38:18,191
这一款版本是现在的预线

788
00:38:18,191 --> 00:38:20,191
这是很新鲜

789
00:38:20,191 --> 00:38:47,611
基本上我认为试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试试�