baalajimaestro/ollama
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
3272 commits 1 branch 0 tags 14 MiB
Commit graph

602 commits

Author SHA1 Message Date
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
Daniel Hiltgen
96624aa412
Merge pull request #5072 from dhiltgen/windows_path 
Move libraries out of users path
 2024-06-19 09:13:39 -07:00
Michael Yang
e873841cbb deepseek v2 graph 2024-06-18 15:35:12 -07:00
Daniel Hiltgen
359b15a597 Handle models with divergent layer sizes 
The recent refactoring of the memory prediction assumed all layers
are the same size, but for some models (like deepseek-coder-v2) this
is not the case, so our predictions were significantly off.
 2024-06-18 11:05:34 -07:00
Daniel Hiltgen
7784ca33ce Tighten up memory prediction logging 
Prior to this change, we logged the memory prediction multiple times
as the scheduler iterates to find a suitable configuration, which can be
confusing since only the last log before the server starts is actually valid.
This now logs once just before starting the server on the final configuration.
It also reports what library instead of always saying "offloading to gpu" when
using CPU.
 2024-06-18 09:15:35 -07:00
Daniel Hiltgen
171796791f Adjust mmap logic for cuda windows for faster model load 
On Windows, recent llama.cpp changes make mmap slower in most
cases, so default to off.  This also implements a tri-state for
use_mmap so we can detect the difference between a user provided
value of true/false, or unspecified.
 2024-06-17 16:54:30 -07:00
Daniel Hiltgen
b0930626c5 Add back lower level parallel flags 
nvcc supports parallelism (threads) and cmake + make can use -j,
while msbuild requires /p:CL_MPcount=8
 2024-06-17 13:44:46 -07:00
Daniel Hiltgen
e890be4814 Revert "More parallelism on windows generate" 
This reverts commit 0577af98f4.
 2024-06-17 13:32:46 -07:00
Daniel Hiltgen
b2799f111b Move libraries out of users path 
We update the PATH on windows to get the CLI mapped, but this has
an unintended side effect of causing other apps that may use our bundled
DLLs to get terminated when we upgrade.
 2024-06-17 13:12:18 -07:00
Jeffrey Morgan
152fc202f5
llm: update llama.cpp commit to 7c26775 (#4896) 
* llm: update llama.cpp submodule to `7c26775`

* disable `LLAMA_BLAS` for now

* `-DLLAMA_OPENMP=off`
 2024-06-17 15:56:16 -04:00
Daniel Hiltgen
4b0050cf0e
Merge pull request #5037 from dhiltgen/faster_win_build 
More parallelism on windows generate
 2024-06-15 08:03:05 -07:00
Daniel Hiltgen
0577af98f4 More parallelism on windows generate 
Make the build faster
 2024-06-15 07:44:55 -07:00
Daniel Hiltgen
da3bf23354 Workaround gfx900 SDMA bugs 
Implement support for GPU env var workarounds, and leverage
this for the Vega RX 56 which needs
HSA_ENABLE_SDMA=0 set to work properly
 2024-06-14 15:38:13 -07:00
Daniel Hiltgen
17df6520c8 Remove mmap related output calc logic 2024-06-14 14:55:50 -07:00
Daniel Hiltgen
6f351bf586 review comments and coverage 2024-06-14 14:55:50 -07:00
Daniel Hiltgen
fc37c192ae Refine CPU load behavior with system memory visibility 2024-06-14 14:51:40 -07:00
Daniel Hiltgen
6fd04ca922 Improve multi-gpu handling at the limit 
Still not complete, needs some refinement to our prediction to understand the
discrete GPUs available space so we can see how many layers fit in each one
since we can't split one layer across multiple GPUs we can't treat free space
as one logical block
 2024-06-14 14:51:40 -07:00
Daniel Hiltgen
fb9cdfa723 Fix server.cpp for the new cuda build macros 2024-06-14 14:51:40 -07:00
Michael Yang
217f60c3d9
Merge pull request #4987 from ollama/mxyng/revert-byte-order 
Revert "Merge pull request #4938 from ollama/mxyng/fix-byte-order"
 2024-06-11 16:04:20 -07:00
Michael Yang
7bdcd1da94 Revert "Merge pull request #4938 from ollama/mxyng/fix-byte-order" 
This reverts commit f5f245cc15, reversing
changes made to 94d37fdcae.

this change broke gguf v2 which is incorrectly detected as big endian
 2024-06-11 15:56:17 -07:00
Jeffrey Morgan
ead259d877
llm: fix seed value not being applied to requests (#4986) 2024-06-11 14:24:41 -07:00
Michael Yang
f5f245cc15
Merge pull request #4938 from ollama/mxyng/fix-byte-order 
fix parsing big endian gguf
 2024-06-10 09:38:12 -07:00
Craig Hughes
b84aea1685
Critical fix from llama.cpp JSON grammar to forbid un-escaped escape characters inside strings, which breaks parsing. (#3782) 2024-06-09 10:57:09 -07:00
Jeffrey Morgan
34f142797a
llm: always add bos token to prompt (#4941) 
* fix embedding by adding fixes from llama.cpp upstream

* remove assert

---------

Co-authored-by: Jesper Ek <deadbeef84@gmail.com>
 2024-06-08 18:47:10 -07:00
Michael Yang
620d5c569e fix parsing big endian gguf 2024-06-08 12:35:26 -07:00
Daniel Hiltgen
cddc63381c
Merge pull request #4909 from dhiltgen/oneapi_disable 
Add ability to skip oneapi generate
 2024-06-07 14:07:15 -07:00
Michael Yang
030e765e76 fix create model when template detection errors 2024-06-07 10:51:35 -07:00
Daniel Hiltgen
ab8c929e20 Add ability to skip oneapi generate 
This follows the same pattern for cuda and rocm to allow
disabling the build even when we detect the dependent libraries
 2024-06-07 08:32:49 -07:00
Jeffrey Morgan
ce0dc33cb8
llm: patch to fix qwen 2 temporarily on nvidia (#4897) 2024-06-06 23:14:33 -07:00
Michael Yang
9b6c2e6eb6 detect chat template from KV 2024-06-06 16:03:47 -07:00
Michael Yang
6297f85606 gofmt, goimports 2024-06-04 13:20:24 -07:00
Michael Yang
e40145a39d lint 2024-06-04 11:13:30 -07:00
Michael Yang
c895a7d13f some gocritic 2024-06-04 11:13:30 -07:00
Michael Yang
04f3c12bb7 replace x/exp/slices with slices 2024-06-04 11:13:30 -07:00
Michael Yang
829ff87bd1
revert tokenize ffi (#4761) 
* Revert "use `int32_t` for call to tokenize (#4738)"

This reverts commit 763bb65dbb.

* Revert "vocab only"

This reverts commit bf54c845e9.

* Revert "use ffi for tokenizing/detokenizing"

This reverts commit 26a00a0410.
 2024-05-31 18:54:21 -07:00
Jeffrey Morgan
763bb65dbb
use int32_t for call to tokenize (#4738) 
* use `int32_t` for call to tokenize

* variable naming

* cleanup

* fix crash
 2024-05-30 21:43:30 -07:00
Jeffrey Morgan
7ca9605f54
speed up tests by only building static lib (#4740) 2024-05-30 21:43:15 -07:00
Michael Yang
eb2c443a79
Merge pull request #4736 from ollama/mxyng/vocab-only 
vocab only for tokenize
 2024-05-30 17:21:00 -07:00
Jeffrey Morgan
a50a87a7b8
partial offloading: allow flash attention and disable mmap (#4734) 
* partial offloading: allow flash attention and disable mmap

* allow mmap with num_gpu=0
 2024-05-30 16:58:01 -07:00
Michael Yang
bf54c845e9 vocab only 2024-05-30 16:49:28 -07:00
Jeffrey Morgan
22f5c12ced
Update llama.cpp submodule to 5921b8f0 (#4731) 
* update llama.cpp submodule to `5921b8f089d3b7bda86aac5a66825df6a6c10603`

* add patch
 2024-05-30 16:20:22 -07:00
Michael Yang
de781b37c8 rm unused infill 2024-05-29 11:26:47 -07:00
Michael Yang
3e21799377 rm unused system prompt 2024-05-29 11:26:47 -07:00
Michael Yang
26a00a0410 use ffi for tokenizing/detokenizing 2024-05-29 11:26:47 -07:00
Daniel Hiltgen
646371f56d
Merge pull request #3278 from zhewang1-intc/rebase_ollama_main 
Enabling ollama to run on Intel GPUs with SYCL backend
 2024-05-28 16:30:50 -07:00
Daniel Hiltgen
92c81e8117 Give the final model loading more time 
On some systems, 1 minute isn't sufficient to finish the load after it
hits 100% This creates 2 distinct timers, although they're both set to
the same value for now so we can refine the timeouts further.
 2024-05-28 09:08:10 -07:00
Lei Jitang
7487229c34
llm/server.go: Fix 2 minor typos (#4661) 
Signed-off-by: Lei Jitang <leijitang@outlook.com>
 2024-05-27 17:21:10 -07:00
Daniel Hiltgen
0165ba1651
Merge pull request #4638 from dhiltgen/better_error 
Report better warning on client closed abort of load
 2024-05-25 14:32:28 -07:00
Daniel Hiltgen
c4209d6d21 Report better warning on client closed abort of load 
If the client closes the connection before we finish loading the model
we abort, so lets make the log message clearer why to help users
understand this failure mode
 2024-05-25 09:23:28 -07:00
Michael Yang
d51f15257c
Update llm/ggml.go 
Co-authored-by: Bruce MacDonald <brucewmacdonald@gmail.com>
 2024-05-24 16:10:43 -07:00
Michael Yang
8f440d579a fix q5_0, q5_1 2024-05-24 16:01:46 -07:00
Patrick Devine
4cc3be3035
Move envconfig and consolidate env vars (#4608) 2024-05-24 14:57:15 -07:00
Wang,Zhe
fd5971be0b support ollama run on Intel GPUs 2024-05-24 11:18:27 +08:00
Michael Yang
714adb8bd1
bump (#4597) 2024-05-23 14:16:26 -07:00
Daniel Hiltgen
95b1133d0c
Merge pull request #4547 from dhiltgen/load_progress 
Wire up load progress
 2024-05-23 14:06:02 -07:00
Daniel Hiltgen
b37b496a12 Wire up load progress 
This doesn't expose a UX yet, but wires the initial server portion
of progress reporting during load
 2024-05-23 13:36:48 -07:00
Bruce MacDonald
d6f692ad1a
Add support for IQ1_S, IQ3_S, IQ2_S, IQ4_XS. IQ4_NL (#4322) 
Co-authored-by: ManniX-ITA <20623405+mann1x@users.noreply.github.com>
 2024-05-23 13:21:49 -07:00
Jeffrey Morgan
38255d2af1
Use flash attention flag for now (#4580) 
* put flash attention behind flag for now

* add test

* remove print

* up timeout for sheduler tests
 2024-05-22 21:52:09 -07:00
Michael Yang
171eb040fc simplify safetensors reading 2024-05-21 11:28:22 -07:00
Michael Yang
bbbd9f20f3 cleanup 2024-05-20 16:13:57 -07:00
Michael Yang
547132e820 bpe pretokenizer 2024-05-20 16:13:57 -07:00
Patrick Devine
c8cf0d94ed llama3 conversion 2024-05-20 16:13:57 -07:00
jmorganca
5cab13739e set llama.cpp submodule commit to 614d3b9 2024-05-20 15:28:17 -07:00
Josh Yan
8aadad9c72 updated updateURL 2024-05-20 15:24:32 -07:00
Sam
e15307fdf4
feat: add support for flash_attn (#4120) 
* feat: enable flash attention if supported

* feat: enable flash attention if supported

* feat: enable flash attention if supported

* feat: add flash_attn support
 2024-05-20 13:36:03 -07:00
Jeffrey Morgan
583c1f472c
update llama.cpp submodule to 614d3b9 (#4414) 2024-05-16 13:53:09 -07:00
Daniel Hiltgen
c48c1d7c46 Port cuda/rocm skip build vars to linux 
Windows already implements these, carry over to linux.
 2024-05-15 15:56:43 -07:00
Patrick Devine
d1692fd3e0
fix the cpu estimatedTotal memory + get the expiry time for loading models (#4461) 2024-05-15 15:43:16 -07:00
Daniel Hiltgen
853ae490e1 Sanitize the env var debug log 
Only dump env vars we care about in the logs
 2024-05-15 14:42:57 -07:00
Michael Yang
0e331c7168
Merge pull request #4328 from ollama/mxyng/mem 
count memory up to NumGPU if set by user
 2024-05-14 13:47:44 -07:00
Patrick Devine
6845988807
Ollama ps command for showing currently loaded models (#4327) 2024-05-13 17:17:36 -07:00
Michael Yang
1d359e737e typo 2024-05-13 14:18:34 -07:00
Michael Yang
50b9056e09 count memory up to NumGPU 2024-05-13 14:13:10 -07:00
jmorganca
92ca2cca95 Revert "only forward some env vars" 
This reverts commit ce3b212d12.
 2024-05-10 22:53:21 -07:00
Daniel Hiltgen
c4014e73a2 Fall back to CPU runner with zero layers 2024-05-10 15:09:48 -07:00
Michael Yang
1eb382da5a add phi2 mem 2024-05-10 12:13:28 -07:00
Jeffrey Morgan
bb6fd02298
Don't clamp ctx size in PredictServerFit (#4317) 
* dont clamp ctx size in `PredictServerFit`

* minimum 4 context

* remove context warning
 2024-05-10 10:17:12 -07:00