jordan (@jmil3000) header
jordan (@jmil3000) profile picture

ml + everything else

山中
Joined July 2020
Photo by jordan (@jmil3000)Photo by jordan (@jmil3000)Photo by jordan (@jmil3000)Photo by jordan (@jmil3000)Photo by jordan (@jmil3000)Photo by jordan (@jmil3000)Photo by jordan (@jmil3000)Photo by jordan (@jmil3000)Photo by jordan (@jmil3000)Photo by jordan (@jmil3000)
been testing some "mini models" on my 3070 (8gb vram). had another to run through my benchmark today, the new bonsai 2 9b. however, this time I sat downstairs playing piano while i prompted free deepseek v4.1f via opencode through codync to complete the bench mark. it was great..
1
1
150
do the new openai models still say "thats not x, its y"
14
jordan retweeted
That nigga gave us the 3070 and you want me to fucking condemn him. Fuck you
48
3,066
78
31,387
536,454
the one guy that had a little bit of aura in ai is now gone, it's so over
21
10,500 gb300s for RL post train, 8,500 of those for inference. crazy!!
Beam’s model technical report is now available. Read the full report to learn about the methodology our team used to train Beam end-to-end in-house, here: reflection.ai/documents/beam…
56
so peak RAHHH
13
the "follow the agent" feature in Zed is so cool
14
The sort of thing your agent writes tests for
The chances of being killed by a landmine delivered by a bear are low, but they are never zero.
77
835
36
14,204
328,663
bro if you regularly use codex in the desktop app (me) try it in Zed. doesn't even feel like the same model its so fast
15
anybody ever tried herdr? thoughts?
13
very nice, added to the reading list
Your agent solved the task at xhigh. Would low have solved it too? You can’t tell without rerunning it, and you still pay for every token. We tested models x reasoning effort level x benchmarks. On tasks low effort can already solve, top effort costs up to 42× more #LLM #AI 🧵👇
2
32
they're calling it some of the best bait of all time
This quoted post is unavailable.
13
I know its random but they done botched my boy 5.6 sol!! he's so dumb now 💔
14
welp, that was the last game GT had a good chance at winning. until next year! Hope it’s better
1
16
while I was busy kimi passed astra? wow
15
todays iteration of this tweet: did the trending "redo my tv ui" via codex while I did the dishes and listened to the cameron winter live solo album. its faster and no ad slop every time I turn on the tv now. not to mention a ui that is built on actually good hci principles
been testing some "mini models" on my 3070 (8gb vram). had another to run through my benchmark today, the new bonsai 2 9b. however, this time I sat downstairs playing piano while i prompted free deepseek v4.1f via opencode through codync to complete the bench mark. it was great..
63
jordan retweeted
this prompt engineering shit is too easy
7
10
117
1,539
Taylor Swift's silence on the PHP CVE is deafening
106
1,658
102
20,471
345,899
man ...
No 5 hour limit is soooo nice
1
25
and this, to me, is what ai should be. I still get done what I need to while freeing up time to create art, or do anything really. I hope future advancements make this the norm more and more. yippee!
1
12