jmorganca
|
9b5b69c00f
llm: update llama.cpp submodule to `7c26775`
|
vor 10 Monaten |
Jeffrey Morgan
|
163cd3e77c
gpu: add env var for detecting Intel oneapi gpus (#5076)
|
vor 10 Monaten |
Daniel Hiltgen
|
4c2c8f93dd
Merge pull request #5080 from dhiltgen/debug_intel_crash
|
vor 10 Monaten |
Daniel Hiltgen
|
fd1e6e0590
Add some more debugging logs for intel discovery
|
vor 10 Monaten |
royjhan
|
89c79bec8c
Add ModifiedAt Field to /api/show (#5033)
|
vor 10 Monaten |
Jeffrey Morgan
|
c7b77004e3
docs: add missing powershell package to windows development instructions (#5075)
|
vor 10 Monaten |
Daniel Hiltgen
|
07d143f412
Merge pull request #5058 from coolljt0725/fix_build_warning
|
vor 10 Monaten |
Daniel Hiltgen
|
4b0050cf0e
Merge pull request #5037 from dhiltgen/faster_win_build
|
vor 10 Monaten |
Daniel Hiltgen
|
0577af98f4
More parallelism on windows generate
|
vor 10 Monaten |
Daniel Hiltgen
|
17ce203a26
Merge pull request #4875 from dhiltgen/rocm_gfx900_workaround
|
vor 10 Monaten |
Daniel Hiltgen
|
d76555ffb5
Merge pull request #4874 from dhiltgen/rocm_v6_bump
|
vor 10 Monaten |
Daniel Hiltgen
|
2786dff5d3
Merge pull request #4264 from dhiltgen/show_gpu_visible_settings
|
vor 10 Monaten |
Lei Jitang
|
225f0d1219
gpu: Fix build warning
|
vor 10 Monaten |
Daniel Hiltgen
|
532db58311
Merge pull request #4972 from jayson-cloude/main
|
vor 10 Monaten |
Daniel Hiltgen
|
6be309e1bd
Centralize GPU configuration vars
|
vor 11 Monaten |
Daniel Hiltgen
|
da3bf23354
Workaround gfx900 SDMA bugs
|
vor 11 Monaten |
Daniel Hiltgen
|
26ab67732b
Bump ROCm linux to 6.1.1
|
vor 10 Monaten |
Daniel Hiltgen
|
45cacbaf05
Merge pull request #4517 from dhiltgen/gpu_incremental
|
vor 10 Monaten |
Daniel Hiltgen
|
17df6520c8
Remove mmap related output calc logic
|
vor 10 Monaten |
Daniel Hiltgen
|
6f351bf586
review comments and coverage
|
vor 11 Monaten |
Daniel Hiltgen
|
ff4f0cbd1d
Prevent multiple concurrent loads on the same gpus
|
vor 11 Monaten |
Daniel Hiltgen
|
fc37c192ae
Refine CPU load behavior with system memory visibility
|
vor 11 Monaten |
Daniel Hiltgen
|
434dfe30c5
Reintroduce nvidia nvml library for windows
|
vor 11 Monaten |
Daniel Hiltgen
|
4e2b7e181d
Refactor intel gpu discovery
|
vor 11 Monaten |
Daniel Hiltgen
|
48702dd149
Harden unload for empty runners
|
vor 11 Monaten |
Daniel Hiltgen
|
68dfc6236a
refined test timing
|
vor 11 Monaten |
Daniel Hiltgen
|
5e8ff556cb
Support forced spreading for multi GPU
|
vor 11 Monaten |
Daniel Hiltgen
|
6fd04ca922
Improve multi-gpu handling at the limit
|
vor 11 Monaten |
Daniel Hiltgen
|
206797bda4
Fix concurrency integration test to work locally
|
vor 11 Monaten |
Daniel Hiltgen
|
43ed358f9a
Refine GPU discovery to bootstrap once
|
vor 11 Monaten |