nulltron

83 posts

nulltron banner
nulltron

nulltron

@nulltron_

experimenting and pursuing curiosity from first principles. occasional quips and blunders.

localhost शामिल हुए Mayıs 2026
16 फ़ॉलोइंग23 फ़ॉलोवर्स
पिन किया गया ट्वीट
nulltron
nulltron@nulltron_·
~jrnl 001 today i had codex grind through my rust yaml library against the leading oracles. thought we were ahead. turns out my fast lane was still disagreeing with my correct lane. codex spent 5 hours grinding out the ~30 cases where we had contradictions. i actually hate yaml. but having a faster lib than saphyr and more correct than ruamel is just a flex --even if it's a slight lead. don tell the normies what im working on. next target: regex.
GIF
English
1
0
3
120
nulltron
nulltron@nulltron_·
@vedad_taranin yeah its not so bad with /goal but you need a plan of attack. you using agents?
English
0
0
0
5
nulltron
nulltron@nulltron_·
~jrnl 001 today i had codex grind through my rust yaml library against the leading oracles. thought we were ahead. turns out my fast lane was still disagreeing with my correct lane. codex spent 5 hours grinding out the ~30 cases where we had contradictions. i actually hate yaml. but having a faster lib than saphyr and more correct than ruamel is just a flex --even if it's a slight lead. don tell the normies what im working on. next target: regex.
GIF
English
1
0
3
120
nulltron
nulltron@nulltron_·
@nutlope my question is does it run locally? if i have to use cloud for it, i'm not that keen
English
0
0
0
374
Hassan
Hassan@nutlope·
This model is insane at design. I asked GLM 5.2 (left) and Opus 4.8 (right) to build me a landing page and you can't even tell the difference. GLM cost $0.06 while opus cost $0.49. More than 6x cheaper while being faster + more token efficient. Another win for open source AI.
Z.ai@Zai_org

Introducing GLM-5.2: Frontier Intelligence, Open Weights - Significant improvements in coding and agentic tasks - Strong long-horizon capabilities with a 1M context window - Two levels of reasoning effort: GLM-5.2 (max) pushes the limits, while GLM-5.2 (high) strikes a strong balance between performance and token efficiency - MIT-licensed open weights - Same API pricing as GLM-5.1 Tech Blog: z.ai/blog/glm-5.2 Weights: huggingface.co/zai-org/GLM-5.2 API: docs.z.ai/guides/llm/glm… Coding Plan: z.ai/subscribe Chat: chat.z.ai

English
85
81
1.4K
170.9K
nulltron
nulltron@nulltron_·
@kimmonismus wait is this self-hostable or a mega model?
English
0
0
0
26
nulltron
nulltron@nulltron_·
@thsottiaux you guys are gonna owe me 2 resets because the first one you said was coming never materialized 🥲
English
1
0
6
1.3K
Nicole
Nicole@karolcoimbra3·
@nulltron_ @OpenAIDevs @rust_foundation OpenAI becoming a Platinum Member of the Rust Foundation was indeed a bit unexpected. Do you think they have any ulterior motives behind doing this?
English
1
0
0
20
nulltron
nulltron@nulltron_·
@danliu yes I think now that Im seeing it everywhere its starting to grow on me. the era of neon green and purple tailwind design I hope is over.
English
0
0
0
392
Dan Liu
Dan Liu@danliu·
there should be a term for this "beige" phase of web design.
Dan Liu tweet media
English
58
2
435
31.4K
nulltron
nulltron@nulltron_·
@RaminNasibov wait, you guys are still touching it? i just scream at it.
English
0
0
0
42
nulltron
nulltron@nulltron_·
my love affair with Strings has ended.
English
0
0
0
18
Andi
Andi@andi_losing·
i’m stopping daily videos here's the reason:
English
29
1
75
3.6K
JARIN 🌊
JARIN 🌊@Web3_Jarin·
Build your X Account Just say "hello " ✌ gain 350+ mutuals under this post
JARIN 🌊 tweet media
English
6K
228
3K
615.2K
Eliana
Eliana@eliana_jordan·
what are you building right now? 1. app 2. saas
English
216
2
107
11.3K
Pascio
Pascio@IAmPascio·
founders, plug yourself what are you building?
English
125
0
47
6.1K
Sahil Panhotra | Indie Builder | Dev
woke up to 90 new followers every "Nobody is seeing my Posts" phase has an expiration date most people quit before it arrives
Sahil Panhotra | Indie Builder | Dev tweet media
English
89
0
127
3K
AETHER
AETHER@aether_uiux·
Be honest, which AI model is actually best for Coding ?
AETHER tweet mediaAETHER tweet media
English
29
1
22
2.1K
nulltron
nulltron@nulltron_·
@rauchg spelled *sleep* wrong
English
0
0
0
13
Guillermo Rauch
Guillermo Rauch@rauchg·
It’s time to ship
English
60
26
508
29.4K
nulltron
nulltron@nulltron_·
@shafu0x im so good i only talk to myself
English
0
0
3
894
shafu
shafu@shafu0x·
be so fucking good you never to talk to recruiters
English
128
509
11.4K
642K