Dhavan

[Dhavan @codingquark]

I hear GPUs are also useful for games 👀

[Dhavan @codingquark]

@0xSero Is it worth saving on RTX 6000 and go with DGX? For agents, tuning, experimenting.

Been renting 6000 and debating if mem bandwidth is okay trade off

[Dhavan @codingquark]

@bryan_johnson Many of us in very hot climates need no convincing. Keep up the great work!

[Dhavan @codingquark]

@wbdu @The_Only_Signal Happens to me once in a while with Qwen3-coder-next. Not so often that I feel frustrated.

[Dhavan @codingquark]

@scaling01 Combine, apply to physics

[Dhavan @codingquark]

Qwen3-coder-next seems like a nice, useful model!

[Dhavan @codingquark]

@Michaelzsguo I just started playing with it on runpod. It stopped midway in opencode as well. Running quantised model btw

[Dhavan @codingquark]

@ether_fan @antirez That’s about right. Openrouter usually gives 25-35

[Dhavan @codingquark]

@esrtweet Nice! Thanks!

[Elon Musk @elonmusk]

The latest 𝕏 algorithm has been published to GitHub
https://t.co/ZCOm51uxmh

[Dhavan @codingquark]

With apps like Superwhisper and Wispr Flow, imagine having a subscription for your keyboard. You want to backspace many words? Sure! Ten bucks!

[Dhavan @codingquark]

I have been playing too much with different models. It is now confusing and yet illuminating how different models behave for different uses

[Dhavan @codingquark]

looks like I won't have to pay for commit messages at least :P

[Dhavan @codingquark]

Revived my old 2080ti, loaded Gemma 3 12B with llama.cpp. This is where we are at:

prompt eval time = 76.59 ms / 17 tokens ( 4.51 ms per token, 221.96 tokens per second)
eval time = 32043.16 ms / 1598 tokens ( 20.05 ms per token, 49.87 tokens per second)
total time = 32119.75 ms / 1615 tokens

[Dhavan @codingquark]

@_HooJR @antirez Totally! Great stuff here

[Dhavan @codingquark]

@Sentdex GM!

[Dhavan @codingquark]

https://t.co/oQh5ocMH76

[Dhavan @codingquark]

It cost me ~US$1 on OpenRouter to fetch latest from meshtastic upstream, compile and flash. GLM-5.1. it made no errors which was great.

[Dhavan @codingquark]

Done with things for now