-
### What happened?
Observed performance drop for quantized kv cache with flash attention. Both pp and tg are 1/3 slower when flash attention enable and one of k,v is quantized.
Here are some bench…
-
I know this states support up to Sonoma, but it looks like Apple have moved the required data files again under Sequoia.
Running /usr/bin/security find-generic-password -l BeaconStore -g
returns
s…
-
Android-auto-companion-ios is not working in Xcode 16 alone.
After adding Android-auto-companion-ios package into our existing iOS Applications the app is not running in Xcode 16 alone, Xcode 15.x …
-
Gravatar supports some cache control headers. Let's investigate and experiment on different cache policies that are defined in Apple's high level networking frameworks and see how we can offer other c…
-
The [APNSwift](https://swiftpackageindex.com/swift-server-community/APNSwift) package is reporting 1 Swift 6 error via `Sema.NumSwift6Errors` while the package itself does not show any concurrency war…
-
### What happened?
I attempted to purchase an application using ipatool, but the purchase failed with the error message as seen below.
I have aswell made several attempts on different syst…
-
Apple figured out that precomputing a {Selector, IMP} static hash table significantly improves dispatch time and the memory footprint. This cache is stored inside the DYLD shared cache and includes th…
-
### Verification
- [X] I searched for recent similar issues at https://github.com/rust-lang/rustup/issues?q=is%3Aissue+is%3Aopen%2Cclosed and found no duplicates.
- [X] I am on the latest version of …
-
Now bear with me here - I know this isn't the repo for the Airplay plugin, but I believe the change needs to be made in this repository, or in the build repository.
When I click "pair" to add a PIN…
-
### What happened?
When using the llama.cpp server with cache_prompt enabled, I've encountered an issue where the logit_bias specified in one request persists and influences subsequent requests, ev…