characharm

After nearly 10 years, I still have to resort to workarounds that unfortunately require a lot of unnecessary "hacks". Why is that? Please, Valve.

I can't say if it's related to the topic of discussion, but the same error is spamming in the console, and some players complain that their teammates are lagging, specifically...

[The authentication response was rejected because the anti-forgery token was invalid.](https://github.com/aspnet-contrib/AspNet.Security.OpenId.Providers/issues/95)

> If I remember correctly this happens when the site is not opened via HTTPS even in localhost?

It seems that llama stopped crashing after b3828, but now the computer completely freezes when loading the model. Applying the patch doesn't change the behavior. Intel Arc A770, latest driver,...

> since it is windows, I suppose you install the drivers all with Intel Arc Control, right? @characharm @easyfab Happy belated holiday to you :) and yes, arc control is...

> I let @characharm confirm that applyig `git revert e62e9789cda3bf5573a747e55ec2a7ee32908f5` over master also solves his freezing Works fine with smaller models, but if the model size is larger (doesn't fit...

@Sketchfellow I tried it, but the llama-server process still gradually eats up memory. The GPU constantly offloads something into system memory as tokens are generated, and since reasoning models generate...

I have a similar issue with llama-server. The launch flags are --jinja -c 15000 -ngl 99 --port 8000, nothing special in the parameters, so at first I thought the problem...

It turns out my problem is most likely not related to this bug. I made a small SYCL utility that constantly pings a small area of memory on the Intel...