AdamLAG

Data Engineer watching AI cook
Also a reader, writer, anime lover, artist... yeah too many hobbies

[AdamLAG @Adamlags]

hello

i m not a ceo of a 4T$ company

i m just some guy who reads AI all day, builds with it, and posts about it

what you get if you follow:

leaks/news i’ve checked
what i learned about AI today
my opinions

like that? follow. if not, also follow lol.

[John Ternus @johnternus]

hello

[AdamLAG @Adamlags]

@theo Gpt 6 luna

[AdamLAG @Adamlags]

@theo Me 2

[AdamLAG @Adamlags]

@thsottiaux @DavidOndrej1 The most hyped reply I saw in my day

[AdamLAG @Adamlags]

@xikhar 3rd one !

[AdamLAG @Adamlags]

@thsottiaux This is the 3rd one after astra !!

[AdamLAG @Adamlags]

@adonis_singh Absolute AI

[AdamLAG @Adamlags]

Astra seems to use way less tokens, take much less time to fi ish tge task, way cheaper than fable 5.1.. gpt 6 is a beast 🔥

[thehype. @thehypedotnews]

gpt astra vs fable 5.1 at goldberg machine

gpt 6 astra – openai, landed on @OpenRouter less then hour ago, provider pinned to openai
fable 5.1 – anthropic, shipped sep 1

we put the two models on one job: a rube goldberg machine in three.js that presses a button and detonates a bomb

the setup: one self-contained html file, three.js from a cdn, everything else procedural – no textures, no models, no physics engine, every collision hand-written. the hard part sits in the brief: a domino may only fall once the previous one actually touches it, checked by real overlap every frame, never by a timer. same rule for the hammer hitting the button and the button firing the bomb. one continuous camera, its speed driven by whatever is moving. we recorded both scenes frame by frame – 1200 frames, 60 fps, exactly 20 seconds – and stepped both by hand to read the telemetry.

- cost
#1 astra – $1.84
#2 fable – $29.16

- time
#1 astra – 9m 56s
#2 fable – 1h 12m

- tokens
#1 astra – 45k
#2 fable – 360k

- lines of code
astra – 881
fable – 744

observations:

• we told it what we saw and nothing else – no diagnosis, no patch. we never edit a model's code. round two ran the whole chain to the blast.

• both files are deterministic. two runs each, identical state to twelve decimals, and neither model reached for math.random.

conclusion: 15.8x cheaper and 7.2x faster, and it still took a second round to get the ball into the bucket!

follow @thehypedotnews for 24/7 ai news, analysis and breakdowns

[AdamLAG @Adamlags]

@thehypedotnews @OpenRouter 1 bucks to 29 is wiiiild

[AdamLAG @Adamlags]

@alexandr_wang Honestly I am impressed I will definitely give muse code a try

[AdamLAG @Adamlags]

@kimmonismus The 6 says it all not like the unusable fable 5.1

[AdamLAG @Adamlags]

@techikansh I cant see it either

[AdamLAG @Adamlags]

@thsottiaux No resets 😭

[AdamLAG @Adamlags]

@ArtificialAnlys Looks like meta did a pretty good job

[AdamLAG @Adamlags]

@ArtificialAnlys Modified it after the astra backlash 🤣🤣

[AdamLAG @Adamlags]

@RoundtableSpace This feels so weird

[AdamLAG @Adamlags]

@tunguz Wait for GPT AGI 1

[AdamLAG @Adamlags]

@alexandr_wang Why aren't you saying gpt who this time huh 🤣

[AdamLAG @Adamlags]

@thsottiaux Master of efficiency.. Anthropic take notes 📝

[AdamLAG @Adamlags]

@kimmonismus Its token price is high but as you said as it is efficient the final price is similar to other low price models that use much more tokens.. this is why cost per task is better than cost per token

[AdamLAG @Adamlags]

@thsottiaux @leploutos Tu parle francais?