Michael Yang
f81f308118
fix gemma, command-r layer weights
2024-04-26 15:00:55 -07:00
Michael Yang
7fea1ecdf6
Merge pull request #3958 from ollama/mxyng/fix-workflow
...
use merge base for diff-tree
2024-04-26 14:17:56 -07:00
Blake Mizerany
054894271d
.github/workflows/test.yaml: add in-flight cancellations on new push ( #3956 )
...
Also, remove a superfluous 'go get'
2024-04-26 13:54:24 -07:00
Michael Yang
6fef042f0b
use merge base for diff-tree
2024-04-26 13:54:15 -07:00
Daniel Hiltgen
5c0c2d1d09
Merge pull request #3954 from dhiltgen/ci_fixes
...
Put back non-avx CPU build for windows
2024-04-26 13:09:03 -07:00
Blake Mizerany
37f9c8ad99
types/model: overhaul Name and Digest types ( #3924 )
2024-04-26 13:08:32 -07:00
Quinten van Buul
2a80f55e2a
Update windows.md ( #3855 )
...
Fixed a typo
2024-04-26 16:04:15 -04:00
Daniel Hiltgen
421c878a2d
Put back non-avx CPU build for windows
2024-04-26 12:44:07 -07:00
Daniel Hiltgen
36666c2142
Merge pull request #3925 from dhiltgen/bump
...
Bump llama.cpp to b2737
2024-04-26 10:09:38 -07:00
Daniel Hiltgen
85801317d1
Fix clip log import
2024-04-26 09:43:46 -07:00
Daniel Hiltgen
2ed0d65948
Bump llama.cpp to b2737
2024-04-26 09:43:28 -07:00
Daniel Hiltgen
d459dc4ad1
Merge pull request #3950 from dhiltgen/windows_packaging
...
Fix exe name for zip packaging on windows
2024-04-26 09:27:37 -07:00
Daniel Hiltgen
40bc4622ef
Fix exe name for zip packaging on windows
...
The zip file encodes the OS and architecture, so keep the short exe name
2024-04-26 09:18:05 -07:00
Daniel Hiltgen
c0f818a07a
Merge pull request #3948 from dhiltgen/win_generate
...
Refactor windows generate for more modular usage
2024-04-26 09:17:20 -07:00
Daniel Hiltgen
8671fdeda6
Refactor windows generate for more modular usage
2024-04-26 08:35:50 -07:00
Daniel Hiltgen
2619850fb4
Merge pull request #3933 from dhiltgen/ci_fixes
...
Move cuda/rocm dependency gathering into generate script
2024-04-26 07:01:24 -07:00
Daniel Hiltgen
8feb97dc0d
Move cuda/rocm dependency gathering into generate script
...
This will make it simpler for CI to accumulate artifacts from prior steps
2024-04-25 22:38:44 -07:00
Daniel Hiltgen
4e1ff6dcbb
Merge pull request #3926 from dhiltgen/ci_fixes
...
Fix release CI
2024-04-25 17:42:31 -07:00
Daniel Hiltgen
8589d752ac
Fix release CI
...
download-artifact path was being used incorrectly. It is where to
extract the zip not the files in the zip to extract. Default is
workspace dir which is what we want, so omit it
2024-04-25 17:27:11 -07:00
Michael Yang
de4ded68b0
Merge pull request #3923 from ollama/mxyng/mem
...
only count output tensors
2024-04-25 16:34:17 -07:00
Daniel Hiltgen
9b5a3c5991
Merge pull request #3914 from dhiltgen/mac_perf
...
Improve mac parallel performance
2024-04-25 16:28:31 -07:00
Jeffrey Morgan
00b0699c75
Reload model if num_gpu changes ( #3920 )
...
* reload model if `num_gpu` changes
* dont reload on -1
* fix tests
2024-04-25 19:02:40 -04:00
Jeffrey Morgan
993cf8bf55
llm: limit generation to 10x context size to avoid run on generations ( #3918 )
...
* llm: limit generation to 10x context size to avoid run on generations
* add comment
* simplify condition statement
2024-04-25 19:02:30 -04:00
Michael Yang
7bb7cb8a60
only count output tensors
2024-04-25 15:24:08 -07:00
Daniel Hiltgen
b123be5b71
Adjust context size for parallelism
2024-04-25 13:58:54 -07:00
jmorganca
ddf5c09a9b
use matrix multiplcation kernels in more cases
2024-04-25 13:58:54 -07:00
Roy Yang
5f73c08729
Remove trailing spaces ( #3889 )
2024-04-25 14:32:26 -04:00
Daniel Hiltgen
f503a848c2
Merge pull request #3895 from brycereitano/shiftloading
...
Move ggml loading to when attempting to fit
2024-04-25 09:24:08 -07:00
Bryce Reitano
36a6daccab
Restructure loading conditional chain
2024-04-24 17:37:03 -06:00
Bryce Reitano
ceb0e26e5e
Provide variable ggml for TestLoad
2024-04-24 17:19:55 -06:00
Bryce Reitano
284e02bed0
Move ggml loading to when we attempt fitting
2024-04-24 17:17:24 -06:00
Michael Yang
3450a57d4a
Merge pull request #3713 from ollama/mxyng/modelname
...
update copy handler to use model.Name
2024-04-24 16:00:32 -07:00
Michael Yang
592dae31c8
update copy to use model.Name
2024-04-24 15:54:54 -07:00
Michael Yang
2010cbc5fa
Merge pull request #3833 from ollama/mxyng/fix-from
...
fix: from blob
2024-04-24 15:13:47 -07:00
Michael Yang
ac0801eced
only replace if it matches command
2024-04-24 14:49:26 -07:00
Michael Yang
ad66e5b060
split temp zip files
2024-04-24 14:18:01 -07:00
Blake Mizerany
ade4b55520
types/model: make ParseName use default without question ( #3886 )
2024-04-24 11:52:55 -07:00
Daniel Hiltgen
a6d62e0617
Merge pull request #3882 from dhiltgen/amd_gfx
...
AMD gfx patch rev is hex
2024-04-24 11:07:49 -07:00
Daniel Hiltgen
6e76348df7
Merge pull request #3834 from dhiltgen/not_found_in_path
...
Report errors on server lookup instead of path lookup failure
2024-04-24 10:50:48 -07:00
Daniel Hiltgen
0d6687f84c
AMD gfx patch rev is hex
...
Correctly handle gfx90a discovery
2024-04-24 09:43:52 -07:00
Patrick Devine
74d2a9ef9a
add OLLAMA_KEEP_ALIVE env variable to FAQ ( #3865 )
2024-04-23 21:06:51 -07:00
Patrick Devine
14476d48cc
fixes for gguf ( #3863 )
2024-04-23 20:57:20 -07:00
Patrick Devine
ce8ce82567
add mixtral 8x7b model conversion ( #3859 )
2024-04-23 20:17:04 -07:00
Blake Mizerany
4dc4f1be34
types/model: restrict digest hash part to a minimum of 2 characters ( #3858 )
...
This allows users of a valid Digest to know it has a minimum of 2
characters in the hash part for use when sharding.
This is a reasonable restriction as the hash part is a SHA256 hash which
is 64 characters long, which is the common hash used. There is no
anticipation of using a hash with less than 2 characters.
Also, add MustParseDigest.
Also, replace Digest.Type with Digest.Split for getting both the type
and hash parts together, which is most the common case when asking for
either.
2024-04-23 18:24:17 -07:00
Daniel Hiltgen
16b52331a4
Merge pull request #3857 from dhiltgen/mem_escape_valve
...
Add back memory escape valve
2024-04-23 17:32:24 -07:00
Daniel Hiltgen
5445aaa94e
Add back memory escape valve
...
If we get our predictions wrong, this can be used to
set a lower memory limit as a workaround. Recent multi-gpu
refactoring accidentally removed it, so this adds it back.
2024-04-23 17:09:02 -07:00
Daniel Hiltgen
2ac3dd6853
Merge pull request #3850 from dhiltgen/windows_packaging
...
Move nested payloads to installer and zip file on windows
2024-04-23 16:35:20 -07:00
Daniel Hiltgen
d8851cb7a0
Harden sched TestLoad
...
Give the go routine a moment to deliver the expired event
2024-04-23 16:14:47 -07:00
Daniel Hiltgen
058f6cd2cc
Move nested payloads to installer and zip file on windows
...
Now that the llm runner is an executable and not just a dll, more users are facing
problems with security policy configurations on windows that prevent users
writing to directories and then executing binaries from the same location.
This change removes payloads from the main executable on windows and shifts them
over to be packaged in the installer and discovered based on the executables location.
This also adds a new zip file for people who want to "roll their own" installation model.
2024-04-23 16:14:47 -07:00
Daniel Hiltgen
790cf34d17
Merge pull request #3846 from dhiltgen/missing_runner
...
Detect and recover if runner removed
2024-04-23 13:14:12 -07:00
