-
**Is your feature request related to a problem? Please describe.**
Relational database systems are not best suited to storing large objects. This can lead to increase runtime costs as well as scalabi…
-
The daily score calculation scales exponentially with the number of models and benchmarks, and is likely the cause of the daily "slow loading" we are currently seeing (previously resulting in regular …
-
While trying to investigate if TAU is also affected by recent issues with CUPTI & OpenMP target offloading, I ran into issues when trying to compile a program with TAU & LLVM when using OpenMP offload…
Thyre updated
6 months ago
-
The natflow already supports IPv4 hwnat, but it seems that it doesn't support IPv6. I noticed that the official Openwrt has added support for IPv6 hardware offloading https://github.com/openwrt/openwr…
-
One of the features that we wanted to add at some point was allowing sbt to hand off all the compilation to Bloop, which would mean that running Bloop in Metals and using sbt CLI would use the same co…
-
Suppose a wallet does not have the compute resources or implementation to generate proofs, but wants to delegate those to a third-party service. What are some protocol options to enable this? There ar…
-
References:
[1] Offloading: https://bisqwit.iki.fi/story/howto/openmp/#OffloadingSupport
[2] Offloading: https://www.lumi-supercomputer.eu/offloading-code-with-compiler-directives/
[3] Very nice …
-
Idea raised in a meeting, spawning an issue here to help keep track of it.
Openmp has an "interop" feature which gives out a handle to the underlying implementation.
I think an interesting optio…
-
I'm a bit lost with the different quantization approaches such as GGUF, ExLlamaV2 & this project?
Is it the same thing? Is one approach faster?
GGUF: [TheBloke/Mixtral-8x7B-v0.1-GGUF](https://hugg…
-
## ❓ Questions and Help
How do you run models that are offloaded to the CPU, Trying to work with ```enable_sequential_cpu_offload``` or ```enable_model_cpu_offload```, when running ```torch_xla.sy…