ZsoltT

6 ms·

1. 1.▲
   by ZsoltT 8mo ago

   > we recommend using SGLang with excess tensor parallelism and EAGLE-3 speculative decoding on live edge Hopper/Blackwell GPUs accessed via low-overhead, prefix-aware HTTP proxies lord
2. 2.▲
   by ZsoltT 8mo ago

   Ahem, so some orgs are adopting a new technology effectively and others aren't. Whether that's because of the nature of their workflows, their organizational effectiveness, or available talent, this sounds like... the way these th
3. 3.▲
   by ZsoltT 10mo ago

   coding was never supposed to be fun, it was supposed to be instrumental you interface with a machine to get the machine to do what you, a human, want it to do relevant to your human purposes but out of necessity - turns out we need to contr
4. 4.▲
   by ZsoltT 1y ago

   glitchless?
5. 5.▲
   by ZsoltT 1y ago

   A correct diagnosis in the context of autoimmune disease. Turned out my severe cognitive symptoms were unrelated to the (until then undiagnosed) autoimmune disease I do, turns out, have GPT5 did the differentials for the complex mishmash of
6. 6.▲
   Opus Outlines the Fate of Man(substack.com)

   2 points by ZsoltT 1y ago|0 comments
7. 7.▲
   OpenAI dev forum is trying to make a little miracle happen(community.openai.com)

   3 points by ZsoltT 3y ago|2 comments
8. 8.▲
   by ZsoltT 3y ago

   Huge org's like OpenAI tend to be pretty opaque to / glacial in addressing small potatoes problems like this one, but they can matter a lot. This one matters a lot. OpenAI does monitor activity on their dev forum but immediately a