- 19 Jan, 2024 8 commits
-
-
Daniel Hiltgen authored
-
Daniel Hiltgen authored
Switch to local dlopen symbols
-
Daniel Hiltgen authored
-
Jeffrey Morgan authored
-
Daniel Hiltgen authored
Fix CPU-only build under Android Termux enviornment.
-
Self Denial authored
-
Self Denial authored
Update gpu.go initGPUHandles() to declare gpuHandles variable before reading it. This resolves an "invalid memory address or nil pointer dereference" error. Update dyn_ext_server.c to avoid setting the RTLD_DEEPBIND flag under __TERMUX__ (Android).
-
Michael Yang authored
fix show handler
-
- 18 Jan, 2024 16 commits
-
-
Michael Yang authored
-
Michael Yang authored
api: add model for all requests
-
Michael Yang authored
ci: update setup-go action
-
Michael Yang authored
-
Michael Yang authored
prefer using req.Model and fallback to req.Name
-
Daniel Hiltgen authored
Mechanical switch from log to slog
-
Michael Yang authored
install: pin fedora to max 37
-
Daniel Hiltgen authored
-
Daniel Hiltgen authored
A few obvious levels were adjusted, but generally everything mapped to "info" level.
-
Mike Bird authored
-
Sachin Sachdeva authored
Updated readme with the web link for haystack ollama integration
-
Daniel Hiltgen authored
Let gpu.go and gen_linux.sh also find CUDA on Arch Linux
-
Daniel Hiltgen authored
Refine the linux cuda/rocm developer docs
-
Daniel Hiltgen authored
Add multiple CPU variants for Intel Mac
-
Daniel Hiltgen authored
-
Daniel Hiltgen authored
The runners are x86 so we can only run binaries that match.
-
- 17 Jan, 2024 4 commits
-
-
Daniel Hiltgen authored
This also refines the build process for the ext_server build.
-
Alexander F. Rødseth authored
-
Michael Yang authored
fix: normalize name path before splitting
-
Michael Yang authored
-
- 16 Jan, 2024 8 commits
-
-
Daniel Hiltgen authored
Bump llama.cpp to b1842 and add new cuda lib dep
-
Daniel Hiltgen authored
Upstream llama.cpp has added a new dependency with the NVIDIA CUDA Driver Libraries (libcuda.so) which is part of the driver distribution, not the general cuda libraries, and is not available as an archive, so we can not statically link it. This may introduce some additional compatibility challenges which we'll need to keep an eye on.
-
Daniel Hiltgen authored
Add macos cross-compile CI coverage
-
Michael Yang authored
repos for fedora 38 and newer do not exist as of this commit ``` $ dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/fedora38/x86_64/cuda-fedora38.repo Adding repo from: https://developer.download.nvidia.com/compute/cuda/repos/fedora38/x86_64/cuda-fedora38.repo Status code: 404 for https://developer.download.nvidia.com/compute/cuda/repos/fedora38/x86_64/cuda-fedora38.repo (IP: 152.195.19.142) Error: Configuration of repo failed ```
-
Michael Yang authored
remove client.py
-
Bruce MacDonald authored
- prompt cache causes inferance to hang after some time
-
Patrick Devine authored
-
Michael Yang authored
fix: request retry with error
-
- 15 Jan, 2024 1 commit
-
-
Daniel Hiltgen authored
improve cuda detection (rel. issue #1704)
-
- 14 Jan, 2024 3 commits
-
-
Daniel Hiltgen authored
-
Daniel Hiltgen authored
Fix typo in arm mac arch script
-
Daniel Hiltgen authored
-