Making computers do things mostly under duress | https://xitter.cf/t.co/HvjV0TyE9E

Joined September 2008
Frankly, the only benchmark that matters nowadays is the "me" benchmark.
57
Once again, I'm noticing how strictly models adhere to system prompts nowadays, they won't budge an inch from the initial constraints
59
EVERYTHING that bundles eMudhra certificates in their release is affected and will need a new release. Easy to spot many "security influencers" that don't know what they are talking about when they treat each CVE(now Java) as something new and try to explain it with mumbo-jumbo.
1
1
127
And for Java the faster way to remediate is this, no need to wait for a new build of the JDK distribution of your choice: ops.java/emudhra (As usual the X Community Notes did a great job "improving" scaremongering tweets)
1
33
Another WebGL example, 6.1 Sol fixing an earlier Argon prototype with correct occlusion, fixed incoming train angle (hard part) and distorted masks. It's way better at this, and with shorted screenshot to fix loop.
電車がきます。
2
1
99
Again, the prompt was just "recreate the scene in this tweet".
8
A few rounds of Gemini 4 Argon starting from the tweet, asking for a Makoto Shinkai aesthetic:
4
A few rounds of Gemini 4 Argon starting from the tweet, asking for a Makoto Shinkai aesthetic:
全てが特別に見えた。
1
1
1
149
Three.js scene, with 2d background with simple 3d models for the card and the train. If you are curious about the one shotted version, here it is:
1
60
There are still a few issues with the mask in the left pole area since it partially disappears when crossed by cars, but many other issues with it were solved iteratively checking screenshot by Argon on it's own, with zero human involvement.
11
And follow @ruiruicamovi for more amazing shots, but that's obvious.
1
それではご覧ください、37年前の日本企業による空飛ぶ自動車を。
386
3,137
589
14,253
699,478
The "I expect people to understand what they are doing" part should also be highlighted.
SCREAM IT FROM THE ROOFTOPS LORD TORVALDS
1
60
Or, to paraphrase: "submissions from meat proxies not appreciated, don't be one."
12
Uh? {"role":"assistant","content":[{"thinking":"**Planning project phases**\n\nI need to think about whether I can realistically manage 27,000 tokens left, especially since the budget is 32,000 and there's only 30,000 available. With 16,000 tokens for time constraints,...
1
97
Since when GPT does estimates of the tokens needed to finish a task and use it as constraint to choose alternative paths?
1
49
"Calculating remaining tokens: I'm noticing we have about 29,000 tokens left. My goal is to manage this effectively for a meaningful extension of ..." This "tokens end" mindset is counterproductive, motivational prompt needed...
30