Daniel Hiltgen
0d651478e4
Merge pull request #3056 from dhiltgen/rocm_link_clash
...
Avoid rocm runner and dependency clash
2024-03-11 09:48:48 -07:00
Michael Yang
9ea492f1ce
convert: fix shape
2024-03-11 09:41:01 -07:00
Daniel Hiltgen
bc13da2bfe
Avoid rocm runner and dependency clash
...
Putting the rocm symlink next to the runners is risky. This moves
the payloads into a subdir to avoid potential clashes.
2024-03-11 09:33:22 -07:00
Jeffrey Morgan
41b00b9856
fix 03-locale.diff
2024-03-10 16:21:05 -07:00
Daniel Hiltgen
c2a8ed48e7
Merge pull request #3048 from dhiltgen/harden_rocm_deps
...
Harden for deps file being empty (or short)
2024-03-10 15:17:22 -07:00
Daniel Hiltgen
3dc1bb6a35
Harden for deps file being empty (or short)
2024-03-10 14:45:38 -07:00
Daniel Hiltgen
7865a6996a
Merge pull request #3046 from dhiltgen/rocm_search_paths
...
Add ollama executable peer dir for rocm
2024-03-10 12:30:56 -07:00
Daniel Hiltgen
00ec269321
Add ollama executable peer dir for rocm
...
This allows people who package up ollama on their own to place
the rocm dependencies in a peer directory to the ollama executable
much like our windows install flow.
2024-03-10 12:16:30 -07:00
Jeffrey Morgan
908005d90b
patch: use default locale in wpm tokenizer ( #3034 )
2024-03-09 21:12:12 -08:00
Jeffrey Morgan
cdf65e793f
only copy deps for amd64 in build_linux.sh
2024-03-09 17:55:22 -08:00
Daniel Hiltgen
82ca694d68
Rename ROCm deps file to avoid confusion ( #3025 )
2024-03-09 17:48:38 -08:00
Jeffrey Morgan
5017a15bcb
add macapp to .dockerignore
2024-03-09 16:07:06 -08:00
Jeffrey Morgan
e11668aa07
add bundle_metal and cleanup_metal funtions to gen_darwin.sh
2024-03-09 16:04:57 -08:00
Jeffrey Morgan
0bd0f4a29c
tidy cleanup logs
2024-03-09 15:56:48 -08:00
Jeffrey Morgan
1ffb1e2874
update llama.cpp submodule to 77d1ac7 ( #3030 )
2024-03-09 15:55:34 -08:00
Daniel Hiltgen
0a7844413c
Merge pull request #3026 from dhiltgen/win_rocm_docs
...
Doc how to set up ROCm builds on windows
2024-03-09 14:17:19 -08:00
Jeffrey Morgan
f9cd55c70b
disable gpu for certain model architectures and fix divide-by-zero on memory estimation
2024-03-09 12:51:38 -08:00
Daniel Hiltgen
0fdebb34a9
Doc how to set up ROCm builds on windows
2024-03-09 11:29:45 -08:00
Daniel Hiltgen
ac64cd4ef9
Merge pull request #3008 from dhiltgen/no_more_idempotent
...
Finish unwinding idempotent payload logic
2024-03-09 09:13:24 -08:00
Daniel Hiltgen
4a5c9b8035
Finish unwinding idempotent payload logic
...
The recent ROCm change partially removed idempotent
payloads, but the ggml-metal.metal file for mac was still
idempotent. This finishes switching to always extract
the payloads, and now that idempotentcy is gone, the
version directory is no longer useful.
2024-03-09 08:34:39 -08:00
Jeffrey Morgan
efe5617b64
update llama.cpp submodule to c2101a2 ( #3020 )
2024-03-09 00:44:50 -08:00
Jeffrey Morgan
5b3fad9636
separate out isLocalIP
2024-03-09 00:22:08 -08:00
Jeffrey Morgan
bfec2c6e10
simplify host checks
2024-03-08 23:29:53 -08:00
Jeffrey Morgan
5c143af726
add additional allowed hosts
2024-03-08 23:23:59 -08:00
Jeffrey Morgan
6c0af2599e
Update docs README.md and table of contents
2024-03-08 22:45:11 -08:00
Jeffrey Morgan
fc8c044584
add allowed host middleware and remove workDir middleware ( #3018 )
2024-03-08 22:23:47 -08:00
Michael Yang
ecc133d843
Merge pull request #3014 from ollama/mxyng/decode-ggla
2024-03-08 16:14:53 -08:00
Michael Yang
76bdebbadf
decode ggla
2024-03-08 15:46:25 -08:00
Michael Yang
18979ad4a1
convert: fix default shape
2024-03-08 15:42:48 -08:00
Michael Yang
8e0ef931d8
Merge pull request #2990 from ollama/mxyng/default-term-size
...
fix: default terminal width, height
2024-03-08 15:20:54 -08:00
Daniel Hiltgen
280da44522
Merge pull request #2988 from dhiltgen/rocm_docs
...
Refined ROCm troubleshooting docs
2024-03-08 13:33:30 -08:00
Bruce MacDonald
0cebc79cba
fix: allow importing a model from name reference ( #3005 )
2024-03-08 12:27:47 -05:00
Jeffrey Morgan
0e4669b04f
update llama.cpp submodule to 6cdabe6 ( #2999 )
2024-03-08 00:26:20 -08:00
Jeffrey Morgan
b886bec3f9
Update api.md
2024-03-07 23:27:51 -08:00
Jeffrey Morgan
fc06205971
Revert "adjust download and upload concurrency based on available bandwidth" ( #2995 )
2024-03-07 18:10:16 -08:00
Blake Mizerany
2ada81e068
cmd: tighten up env var usage sections ( #2962 )
...
Also, document OLLAMA_HOST client semantics per command that honors it.
This looks nicer than having a general puprose environment variable
section in the root usage which was showing up after the "addition help
topics" section outputed by Cobra's default template.
It was decided this was easier to work with than using a custom template
for Cobra right now.
2024-03-07 13:57:07 -08:00
Michael Yang
b1e74d4fda
default terminal width, height
2024-03-07 11:35:42 -08:00
Michael Yang
f678f5c5c3
Merge pull request #2991 from ollama/mxyng/fix-ci
...
fix ci
2024-03-07 11:35:06 -08:00
Michael Yang
2cb74e23fb
fix ci
2024-03-07 11:33:49 -08:00
Daniel Hiltgen
69f0227813
Refined ROCm troubleshooting docs
2024-03-07 11:22:37 -08:00
Daniel Hiltgen
3c8df3808b
Merge pull request #2885 from dhiltgen/rocm_v6_only
...
Revamp ROCm support
2024-03-07 10:51:00 -08:00
Michael Yang
7d564835c2
Merge pull request #2985 from ollama/rm-empty-examples
...
remove empty examples
2024-03-07 10:49:40 -08:00
Michael Yang
72431031d9
no ci test on docs, examples
2024-03-07 10:44:48 -08:00
Michael Yang
6041abb5b2
remove empty examples
2024-03-07 10:40:32 -08:00
Daniel Hiltgen
6c5ccb11f9
Revamp ROCm support
...
This refines where we extract the LLM libraries to by adding a new
OLLAMA_HOME env var, that defaults to `~/.ollama` The logic was already
idempotenent, so this should speed up startups after the first time a
new release is deployed. It also cleans up after itself.
We now build only a single ROCm version (latest major) on both windows
and linux. Given the large size of ROCms tensor files, we split the
dependency out. It's bundled into the installer on windows, and a
separate download on windows. The linux install script is now smart and
detects the presence of AMD GPUs and looks to see if rocm v6 is already
present, and if not, then downloads our dependency tar file.
For Linux discovery, we now use sysfs and check each GPU against what
ROCm supports so we can degrade to CPU gracefully instead of having
llama.cpp+rocm assert/crash on us. For Windows, we now use go's windows
dynamic library loading logic to access the amdhip64.dll APIs to query
the GPU information.
2024-03-07 10:36:50 -08:00
Michael Yang
2e20110e50
Merge pull request #2221 from ollama/mxyng/up-down-ccy
...
adjust download and upload concurrency based on available bandwidth
2024-03-07 09:27:33 -08:00
Daniel Hiltgen
82ddc3e441
Merge pull request #2964 from dhiltgen/mem_limit_var
...
Allow setting max vram for workarounds
2024-03-07 09:25:44 -08:00
Jeffrey Morgan
d481fb3cc8
update go to 1.22 in other places ( #2975 )
2024-03-07 07:39:49 -08:00
DJ Johnson
23ee633252
docs: Add LLM-X to Web Integration section ( #2759 )
2024-03-07 10:11:53 -05:00
John
23ebe8fe11
fix some typos ( #2973 )
...
Signed-off-by: hishope <csqiye@126.com>
2024-03-06 22:50:11 -08:00
