baalajimaestro/ollama
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
3120 commits 1 branch 0 tags 14 MiB
Commit graph

578 commits

Author SHA1 Message Date
Josh
10e768826c
fix: quant err message (#5616) 2024-07-11 17:24:29 -07:00
Jeffrey Morgan
c4cf8ad559
llm: avoid loading model if system memory is too small (#5637) 
* llm: avoid loading model if system memory is too small

* update log

* Instrument swap free space

On linux and windows, expose how much swap space is available
so we can take that into consideration when scheduling models

* use `systemSwapFreeMemory` in check

---------

Co-authored-by: Daniel Hiltgen <daniel@ollama.com>
 2024-07-11 16:42:57 -07:00
Jeffrey Morgan
791650ddef
sched: only error when over-allocating system memory (#5626) 2024-07-11 00:53:12 -07:00
Jeffrey Morgan
efbf41ed81
llm: dont link cuda with compat libs (#5621) 2024-07-10 20:01:52 -07:00
Michael Yang
37a570f962
Merge pull request #5612 from ollama/mxyng/mem 
chatglm graph
 2024-07-10 14:18:33 -07:00
Michael Yang
5a739ff4cb chatglm graph 2024-07-10 13:43:47 -07:00
Jeffrey Morgan
4e262eb2a8
remove GGML_CUDA_FORCE_MMQ=on from build (#5588) 2024-07-10 13:17:13 -07:00
Daniel Hiltgen
b50c818623
Merge pull request #5607 from dhiltgen/win_rocm_v6 
Bump ROCm on windows to 6.1.2
 2024-07-10 12:47:10 -07:00
Daniel Hiltgen
1f50356e8e Bump ROCm on windows to 6.1.2 
This also adjusts our algorithm to favor our bundled ROCm.
I've confirmed VRAM reporting still doesn't work properly so we
can't yet enable concurrency by default.
 2024-07-10 11:01:22 -07:00
Daniel Hiltgen
22c81f62ec Remove duplicate merge glitch 2024-07-10 09:01:33 -07:00
Daniel Hiltgen
2d1e3c3229
Merge pull request #5503 from dhiltgen/dual_rocm 
Workaround broken ROCm p2p copy
 2024-07-09 15:44:16 -07:00
Daniel Hiltgen
b51e3b63ac Statically link c++ and thread lib 
This makes sure we statically link the c++ and thread library on windows
to avoid unnecessary runtime dependencies on non-standard DLLs
 2024-07-09 11:34:30 -07:00
Michael Yang
9bbddc37a7
Merge pull request #5126 from ollama/mxyng/messages 
update message processing
 2024-07-09 09:20:44 -07:00
Daniel Hiltgen
0bacb30007 Workaround broken ROCm p2p copy 
Enable the build flag for llama.cpp to use CPU copy for multi-GPU scenarios.
 2024-07-08 09:40:52 -07:00
Jeffrey Morgan
53da2c6965
llm: remove ambiguous comment when putting upper limit on predictions to avoid infinite generation (#5535) 2024-07-07 14:32:05 -04:00
Jeffrey Morgan
d8def1ff94
llm: allow gemma 2 to context shift (#5534) 2024-07-07 13:41:51 -04:00
Jeffrey Morgan
571dc61955
Update llama.cpp submodule to a8db2a9c (#5530) 2024-07-07 13:03:09 -04:00
Jeffrey Morgan
0e09c380fc
llm: print caching notices in debug only (#5533) 2024-07-07 12:38:04 -04:00
Jeffrey Morgan
4607c70641
llm: add -DBUILD_SHARED_LIBS=off to common cpu cmake flags (#5520) 2024-07-06 18:58:16 -04:00
jmorganca
a08f20d910 release: remove unwanted mingw dll.a files 2024-07-06 15:21:15 -04:00
jmorganca
6cea036027 Revert "llm: only statically link libstdc++" 
This reverts commit 5796bfc401.
 2024-07-06 15:10:48 -04:00
jmorganca
5796bfc401 llm: only statically link libstdc++ 2024-07-06 14:06:20 -04:00
jmorganca
f1a379aa56 llm: statically link pthread and stdc++ dependencies in windows build 2024-07-06 12:54:02 -04:00
jmorganca
9ae146993e llm: add GGML_STATIC flag to windows static lib 2024-07-06 03:27:05 -04:00
Jeffrey Morgan
e0348d3fe8
llm: add COMMON_DARWIN_DEFS to arm static build (#5513) 2024-07-05 22:42:42 -04:00
Jeffrey Morgan
2cc854f8cb
llm: fix missing dylibs by restoring old build behavior on Linux and macOS (#5511) 
* Revert "fix cmake build (#5505)"

This reverts commit 4fd5f3526a.

* llm: fix missing dylibs by restoring old build behavior

* crlf -> lf
 2024-07-05 21:48:31 -04:00
Jeffrey Morgan
5304b765b2
llm: put back old include dir (#5507) 
* llm: put back old include dir

* llm: update link paths for old submodule commits
 2024-07-05 19:34:21 -04:00
Jeffrey Morgan
4fd5f3526a
fix cmake build (#5505) 2024-07-05 19:07:01 -04:00
Michael Yang
ac7a842e55 fix model reloading 
ensure runtime model changes (template, system prompt, messages,
options) are captured on model updates without needing to reload the
server
 2024-07-05 13:17:25 -07:00
Jeffrey Morgan
78fb33dd07
fix typo in cgo directives in llm.go (#5501) 2024-07-05 15:18:36 -04:00
Jeffrey Morgan
8f8e736b13
update llama.cpp submodule to d7fd29f (#5475) 2024-07-05 13:25:58 -04:00
Jeffrey Morgan
d89454de80
Use slot with cached prompt instead of least recently used (#5492) 
* Use common prefix to select slot

* actually report `longest`
 2024-07-05 12:32:47 -04:00
Jeffrey Morgan
e9188e971a
Fix assert on small embedding inputs (#5491) 
* Fix assert on small embedding inputs

* Update llm/patches/09-pooling.diff
 2024-07-05 11:20:57 -04:00
Daniel Hiltgen
02c24d3d01
Merge pull request #5466 from dhiltgen/fix_clip_unicode 
Fix clip model loading with unicode paths
 2024-07-05 08:16:58 -07:00
Jeffrey Morgan
4d71c559b2
fix error detection by limiting model loading error parsing (#5472) 2024-07-03 20:04:30 -04:00
Daniel Hiltgen
ccd7785859
Merge pull request #5243 from dhiltgen/modelfile_use_mmap 
Fix use_mmap for modefiles
 2024-07-03 13:59:42 -07:00
royjhan
3b5a4a77f3
Return Correct Prompt Eval Count Regardless of Cache Prompt (#5371) 
* openai compatibility

* Revert "openai compatibility"

This reverts commit d3f98a811e00fc497d889c8c45b0cfec5b64690c.

* remove erroneous subtraction of prompt cache
 2024-07-03 13:46:23 -07:00
Daniel Hiltgen
0e982bc1f4 Fix corner cases on tmp cleaner on mac 
When ollama is running a long time, tmp cleaners can remove the
runners.  This tightens up a few corner cases on arm macs where
we failed with "server cpu not listed in available servers map[]"
 2024-07-03 13:10:14 -07:00
Daniel Hiltgen
6298f49816 Fix clip model loading with unicode paths 
On windows, if the model dir contained unicode characters
clip models would fail to load.  This fixes the file name
handling in clip.cpp to support utf16 on windows.
 2024-07-03 12:46:36 -07:00
Josh Yan
33a65e3ba3 error 2024-07-01 16:04:13 -07:00
Daniel Hiltgen
97c9e11768 Switch use_mmap to a pointer type 
This uses nil as undefined for a cleaner implementation.
 2024-07-01 08:44:59 -07:00
Daniel Hiltgen
3518aaef33
Merge pull request #4218 from dhiltgen/auto_parallel 
Enable concurrency by default
 2024-07-01 08:32:29 -07:00
Jeffrey Morgan
717f7229eb
Do not shift context for sliding window models (#5368) 
* Do not shift context for sliding window models

* truncate prompt > 2/3 tokens

* only target gemma2
 2024-06-28 19:39:31 -07:00
Michael Yang
de2163dafd gemma2 graph 2024-06-27 13:34:52 -07:00
Jeffrey Morgan
4d311eb731
llm: architecture patch (#5316) 2024-06-26 21:38:12 -07:00
Blake Mizerany
cb42e607c5
llm: speed up gguf decoding by a lot (#5246) 
Previously, some costly things were causing the loading of GGUF files
and their metadata and tensor information to be VERY slow:

  * Too many allocations when decoding strings
  * Hitting disk for each read of each key and value, resulting in a
    not-okay amount of syscalls/disk I/O.

The show API is now down to 33ms from 800ms+ for llama3 on a macbook pro
m3.

This commit also prevents collecting large arrays of values when
decoding GGUFs (if desired). When such keys are encountered, their
values are null, and are encoded as such in JSON.

Also, this fixes a broken test that was not encoding valid GGUF.
 2024-06-24 21:47:52 -07:00
Daniel Hiltgen
17b7186cd7 Enable concurrency by default 
This adjusts our default settings to enable multiple models and parallel
requests to a single model.  Users can still override these by the same
env var settings as before.  Parallel has a direct impact on
num_ctx, which in turn can have a significant impact on small VRAM GPUs
so this change also refines the algorithm so that when parallel is not
explicitly set by the user, we try to find a reasonable default that fits
the model on their GPU(s).  As before, multiple models will only load
concurrently if they fully fit in VRAM.
 2024-06-21 15:45:05 -07:00
Daniel Hiltgen
5bf5aeec01 Refine mmap default logic on linux 
If we try to use mmap when the model is larger than the system free space, loading is slower than the no-mmap approach.
 2024-06-20 11:07:04 -07:00
Michael Yang
8e0641a9bf handle asymmetric embedding KVs 2024-06-20 09:57:27 -07:00
Michael Yang
9d91e5e587 remove confusing log message 2024-06-19 11:14:11 -07:00