evl

@MarcinM02 Share more as you go. I expect qwen to be much worse (just from empiric testing), but I wonder what it looks like dude by side to 5.4.

Got my DGX setup with some Qwen models, and now hooked up to openclaw and hermes agents. Let's see how bad would quality drop vs openAI that I've used before (now I am running Qwen3.5 35B). Running model locally is an amazing feeling !

@MarcinM02 It would be a lie to say I'm not jealous. Quite a piece of hardware!

Got a new toy - Nvidia DGX spark - will try to setup some local models for inference. in the past i used my 3090 (24 gb vram) - but this should allow a lot larger models (128gb).

I guess it's time we learn some Ethereum.

ZKsync prover optimizations just hit mainnet. Big gains on speed & cost. Feel free to read more (technical) details here: amstelden.com/zksync-prover-… DMs open for follow-up questions.

Happy to see many of you interacting with ZKsync's Prover API. I've seen healthy appetite in this direction, but also a lot of struggle on getting the proofs right (plenty of 400s on proof validation). If you're stuck in the process or need help troubleshooting your setup, DM me.