Which model/quant/command line did you use? I can barely get 100 token/secs and for sure clearly not a full context. With vllm, I am limited to 130k tokens with vllm + nvfp4.
1,243 karma · joined October 26, 2018
DeepSeek is okay for random API-based stuff, as it's cheap.
Local open models running on a 5090 are hit or miss. I feel that most GGUFs/quants are awful...
Suddenly, everyone are expert writing the most elegant, clean and without bug code.
Both sucks in French but at least chatgpt prose is readable while Claude is awful.
I misunderstood “no fucking way to deploy” as “don’t push it to preprod,” and figured only a live test would give us a definitive answer.
That one’s on me. I owe you an apology.
On the bright side, Voyager 2 died so Voyager 3 could be future-proof.
Design is too expensive, we do agile. QA too expensive, we fire all of them, and claim devops is the now, which allows us to fire the Ops team too, 100% ownership from deisng to ops on devs.
One person with an agent can replace all these teams. Yeah mo profits.
Most people or even restaurant don't let their meat outside nor use an actual hot pan.
Everything is vibe sloped to death, and dead after a few months to a couple of years (and not hard to be 100 cheaper than GPT-5.6 sol ... DS is basically free and I guess already 100 times cheaper or more, and here another slope ).
But your answer is the first one that seems to be relevant, so thank you.