Commit graph

818 commits

Author SHA1 Message Date
Bach Le
c9c74b4e3f
llama : add classifier-free guidance (#2135)
* Initial implementation

* Remove debug print

* Restore signature of llama_init_from_gpt_params

* Free guidance context

* Make freeing of guidance_ctx conditional

* Make Classifier-Free Guidance a sampling function

* Correct typo. CFG already means context-free grammar.

* Record sampling time in llama_sample_classifier_free_guidance

* Shift all values by the max value before applying logsoftmax

* Fix styling based on review
2023-07-11 19:18:43 +03:00
Jinwoo Jeong
3ec7e596b2
docker : add '--server' option (#2174) 2023-07-11 19:12:35 +03:00
Chad Brewbaker
917831c63a
readme : fix zig build instructions (#2171) 2023-07-11 19:03:06 +03:00
Howard Su
2347463201
Support using mmap when applying LoRA (#2095)
* Support using mmap when applying LoRA

* Fix Linux

* Update comment to reflect the support lora with mmap
2023-07-11 22:37:01 +08:00
LostRuins
bbef28218f
Possible solution to allow K-quants on models with n_vocab!=32000 (#2148)
* This allows LLAMA models that were previously incompatible with K quants to function mostly as normal. This happens when a model has a vocab != 32000, e.g 32001 which means it's not divisible by 256 or 64. Since the problematic dimensions only apply for `tok_embeddings.weight` and `output.weight` (dimentions 4096 x n_vocab), we can simply quantize these layers to Q8_0 whereas the majority of the hidden layers are still K-quanted since they have compatible dimensions.

* Fix indentation

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* As an alternative, to avoid failing on Metal due to lack of Q8_0 support, instead quantize tok_embeddings.weight to Q4_0 and retain output.weight as F16. This results in a net gain of about 55mb for a 7B model compared to previous approach, but should minimize adverse impact to model quality.

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-07-11 22:01:08 +08:00
Evan Miller
5656d10599
mpi : add support for distributed inference via MPI (#2099)
* MPI support, first cut

* fix warnings, update README

* fixes

* wrap includes

* PR comments

* Update CMakeLists.txt

* Add GH workflow, fix test

* Add info to README

* mpi : trying to move more MPI stuff into ggml-mpi (WIP) (#2099)

* mpi : add names for layer inputs + prep ggml_mpi_graph_compute()

* mpi : move all MPI logic into ggml-mpi

Not tested yet

* mpi : various fixes - communication now works but results are wrong

* mpi : fix output tensor after MPI compute (still not working)

* mpi : fix inference

* mpi : minor

* Add OpenMPI to GH action

* [mpi] continue-on-error: true

* mpi : fix after master merge

* [mpi] Link MPI C++ libraries to fix OpenMPI

* tests : fix new llama_backend API

* [mpi] use MPI_INT32_T

* mpi : factor out recv / send in functions and reuse

* mpi : extend API to allow usage with outer backends (e.g. Metal)

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-07-10 18:49:56 +03:00
oobabooga
1d16309969
llama : remove "first token must be BOS" restriction (#2153) 2023-07-09 11:59:53 +03:00
Nigel Bosch
db4047ad5c
main : escape prompt prefix/suffix (#2151) 2023-07-09 11:56:18 +03:00
JackJollimore
18780e0a5e
readme : update Termux instructions (#2147)
The file pathing is significant when running models inside of Termux on Android devices. llama.cpp performance is improved with loading a .bin from the $HOME directory.
2023-07-09 11:20:43 +03:00
clyang
3bbc1a11f0
ggml : fix buidling with Intel MKL but ask for "cblas.h" issue (#2104) (#2115)
* Fix buidling with Intel MKL but ask for "cblas.h" issue

* Use angle brackets to indicate the system library
2023-07-09 11:12:20 +03:00
rankaiyx
2492a53fd0
readme : add more docs indexes (#2127)
* Update README.md to add more docs indexes

* Update README.md to add more docs indexes
2023-07-09 10:38:42 +03:00
Johannes Gäßler
64639555ff
Fixed OpenLLaMA 3b CUDA mul_mat_vec_q (#2144) 2023-07-08 20:01:44 +02:00
Johannes Gäßler
061f5f8d21
CUDA: add __restrict__ to mul mat vec kernels (#2140) 2023-07-08 00:25:15 +02:00
dylan
84525e7962
docker : add support for CUDA in docker (#1461)
Co-authored-by: canardleteer <eris.has.a.dad+github@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-07-07 21:25:25 +03:00
Georgi Gerganov
a7e20edf22
ci : switch threads to 1 (#2138) 2023-07-07 21:23:57 +03:00
Qingyou Meng
1d656d6360
ggml : change ggml_graph_compute() API to not require context (#1999)
* ggml_graph_compute: deprecate using ggml_context, try resolve issue #287

* rewrite: no longer consider backward compitability; plan and make_plan

* minor: rename ctx as plan; const

* remove ggml_graph_compute from tests/test-grad0.c, but current change breaks backward

* add static ggml_graph_compute_sugar()

* minor: update comments

* reusable buffers

* ggml : more consistent naming + metal fixes

* ggml : fix docs

* tests : disable grad / opt + minor naming changes

* ggml : add ggml_graph_compute_with_ctx()

- backwards compatible API
- deduplicates a lot of copy-paste

* ci : enable test-grad0

* examples : factor out plan allocation into a helper function

* llama : factor out plan stuff into a helper function

* ci : fix env

* llama : fix duplicate symbols + refactor example benchmark

* ggml : remove obsolete assert + refactor n_tasks section

* ggml : fix indentation in switch

* llama : avoid unnecessary bool

* ggml : remove comments from source file and match order in header

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-07-07 19:24:01 +03:00
Georgi Gerganov
7242140283 ggml : remove sched_yield() call in ggml_graph_compute_thread() (#2134) 2023-07-07 18:37:10 +03:00
Aarni Koskela
3e08ae99ce
convert.py: add mapping for safetensors bf16 (#1598)
Fixes #1473
2023-07-07 09:12:49 -04:00
Howard Su
481f793acc
Fix opencl by wrap #if-else-endif with \n (#2086) 2023-07-07 05:34:18 +02:00
Georgi Gerganov
dfd9fce6d6
ggml : fix restrict usage 2023-07-06 19:41:31 +03:00
Judd
36680f6e40
convert : update for baichuan (#2081)
1. guess n_layers;
2. relax warnings on context size;
3. add a note that its derivations are also supported.

Co-authored-by: Judd <foldl@boxvest.com>
2023-07-06 19:23:49 +03:00
tslmy
a17a2683d8
alpaca.sh : update model file name (#2074)
The original file name, `ggml-alpaca-7b-q4.bin`, implied the first-generation GGML. After the breaking changes (mentioned in https://github.com/ggerganov/llama.cpp/issues/382), `llama.cpp` requires GGML V3 now. Those model files are named `*ggmlv3*.bin`. We should change the example to an actually working model file, so that this thing is more likely to run out-of-the-box for more people, and less people would waste time downloading the old Alpaca model.
2023-07-06 19:17:50 +03:00
Tobias Lütke
31cfbb1013
Expose generation timings from server & update completions.js (#2116)
* use javascript generators as much cleaner API

Also add ways to access completion as promise and EventSource

* export llama_timings as struct and expose them in server

* update readme, update baked includes

* llama : uniform variable names + struct init

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-07-05 16:51:13 -04:00
Jesse Jojo Johnson
983b555e9d
Update Server Instructions (#2113)
* Update server instructions for web front end
* Update server README
* Remove duplicate OAI instructions
* Fix duplicate text

---------

Co-authored-by: Jesse Johnson <thatguy@jessejojojohnson.com>
2023-07-05 21:03:19 +03:00
Georgi Gerganov
ec326d350c
ggml : fix bug introduced in #1237 2023-07-05 20:44:11 +03:00
Georgi Gerganov
1b6efeab82
tests : fix test-grad0 2023-07-05 20:20:25 +03:00
Stephan Walter
1b107b8550
ggml : generalize quantize_fns for simpler FP16 handling (#1237)
* Generalize quantize_fns for simpler FP16 handling

* Remove call to ggml_cuda_mul_mat_get_wsize

* ci : disable FMA for mac os actions

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-07-05 19:13:06 +03:00
Jesse Jojo Johnson
8567c76b53
Update server instructions for web front end (#2103)
Co-authored-by: Jesse Johnson <thatguy@jessejojojohnson.com>
2023-07-05 18:13:35 +03:00
Johannes Gäßler
924dd22fd3
Quantized dot products for CUDA mul mat vec (#2067) 2023-07-05 14:19:42 +02:00
Howard Su
051c70dcd5
llama: Don't double count the sampling time (#2107) 2023-07-05 18:31:23 +08:00
Johannes Gäßler
9e4475f5cf
Fixed OpenCL offloading prints (#2082) 2023-07-05 08:58:05 +02:00
Nigel Bosch
7f0e9a775e
embd-input: Fix input embedding example unsigned int seed (#2105) 2023-07-05 07:33:33 +08:00
Georgi Gerganov
b472f3fca5
readme : add link web chat PR 2023-07-04 22:25:22 +03:00
Georgi Gerganov
ed9a54e512
ggml : sync latest (new ops, macros, refactoring) (#2106)
- add ggml_argmax()
- add ggml_tanh()
- add ggml_elu()
- refactor ggml_conv_1d() and variants
- refactor ggml_conv_2d() and variants
- add helper macros to reduce code duplication in ggml.c
2023-07-04 21:54:11 +03:00
jwj7140
f257fd2550
Add an API example using server.cpp similar to OAI. (#2009)
* add api_like_OAI.py
* add evaluated token count to server
* add /v1/ endpoints binding
2023-07-04 21:06:12 +03:00
Tobias Lütke
7ee76e45af
Simple webchat for server (#1998)
* expose simple web interface on root domain

* embed index and add --path for choosing static dir

* allow server to multithread

because web browsers send a lot of garbage requests we want the server
to multithread when serving 404s for favicon's etc. To avoid blowing up
llama we just take a mutex when it's invoked.


* let's try this with the xxd tool instead and see if msvc is happier with that

* enable server in Makefiles

* add /completion.js file to make it easy to use the server from js

* slightly nicer css

* rework state management into session, expose historyTemplate to settings

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-07-04 16:05:27 +02:00
Henri Vasserman
acc111caf9
Allow old Make to build server. (#2098)
Also make server build by default.

Tested with Make 3.82
2023-07-04 15:38:04 +03:00
ZhouYuChen
23c7c6fc91
Update Makefile: clean simple (#2097) 2023-07-04 14:15:16 +02:00
Erik Scholz
698efad5fb
CI: make the brew update temporarily optional. (#2092)
until they decide to fix the brew installation in the macos runners.
see the open issues. eg https://github.com/actions/runner-images/pull/7710
2023-07-04 01:50:12 +02:00
Govlzkoy
14a2cc71f6
[ggml] fix index for ne03 value in ggml_cl_mul_f32 (#2088) 2023-07-04 07:50:00 +08:00
Henri Vasserman
1cf14ccef1
fix server crashes (#2076) 2023-07-04 00:05:23 +03:00
Howard Su
cc45a7feb8
Fix crash of test-tokenizer-0 under Debug build (#2064)
* Fix crash of test-tokenizer-0 under Debug build

* Change per comment
2023-07-03 20:43:55 +02:00
Howard Su
55dbb915cc
[llama] No need to check file version when loading vocab score (#2079) 2023-07-03 19:58:58 +08:00
WangHaoranRobin
d7d2e6a0f0
server: add option to output probabilities for completion (#1962)
* server: add option to output probabilities for completion
* server: fix issue when handling probability output for incomplete tokens for multibyte character generation
* server: fix llama_sample_top_k order
* examples/common.h: put all bool variables in gpt_params together
2023-07-03 00:38:44 +03:00
Georgi Gerganov
46088f7231 ggml : fix build with OpenBLAS (close #2066) 2023-07-02 09:46:46 +03:00
Johannes Gäßler
0bc2cdfc87
Better CUDA synchronization logic (#2057) 2023-07-01 21:49:44 +02:00
Johannes Gäßler
befb3a3562
Test-based VRAM scratch size + context adjustment (#2056) 2023-07-01 21:47:26 +02:00
Daniel Drake
b213227067
cmake : don't force -mcpu=native on aarch64 (#2063)
It's currently not possible to cross-compile llama.cpp for aarch64
because CMakeLists.txt forces -mcpu=native for that target.

-mcpu=native doesn't make sense if your build host is not the
target architecture, and clang rejects it for that reason, aborting the
build. This can be easily reproduced using the current Android NDK to build
for aarch64 on an x86_64 host.

If there is not a specific CPU-tuning target for aarch64 then -mcpu
should be omitted completely. I think that makes sense, there is not
enough variance in the aarch64 instruction set to warrant a fixed -mcpu
optimization at this point. And if someone is building natively and wishes
to enable any possible optimizations for the host device, then there is
already the LLAMA_NATIVE option available.

Fixes #495.
2023-07-01 21:31:44 +03:00
Aaron Miller
2f8cd979ec
metal : release buffers when freeing metal context (#2062) 2023-07-01 21:14:59 +03:00
Judd
471aab6e4c
convert : add support of baichuan-7b (#2055)
Co-authored-by: Judd <foldl@boxvest.com>
2023-07-01 20:00:25 +03:00