root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-27 03:44:35 +00:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity
1,466 Commits 359 Branches 2,870 Tags 392 MiB
197a0a9e23
Commit Graph

1466 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Georgi Gerganov
197a0a9e23
server : fix switch fallthrough 2023-10-22 16:55:05 +03:00
Georgi Gerganov
715f384a6b
clip : link to ggml, not to llama 2023-10-22 16:52:12 +03:00
Georgi Gerganov
4b4ab722ab
make : silence stb warnings 2023-10-22 16:51:59 +03:00
Georgi Gerganov
176993c871
Merge branch 'master' into server-rev 2023-10-22 15:04:16 +03:00
Georgi Gerganov
22c69a2794
batched : add len CLI argument 2023-10-22 08:37:20 +03:00
FSSRepo
2eb4c11ec5 fix image load + view image in chat 2023-10-21 14:34:19 -04:00
Jhen-Jie Hong
17b23eb9cb
server : fix multibyte handle in partial response (#3706) 2023-10-21 14:58:03 +03:00
shibe2
465219b914 CLBlast: Add outer loops over src0 for broadcasting in mulmat 
Reduce repeated dequantization of the same data.
 2023-10-20 22:30:52 +04:00
Georgi Gerganov
d1031cf49c
sampling : refactor init to use llama_sampling_params (#3696) 
* sampling : refactor init to use llama_sampling_params

* llama : combine repetition, frequency and presence penalties in 1 call

* examples : remove embd-input and gptneox-wip

* sampling : rename penalty params + reduce size of "prev" vector

* sampling : add llama_sampling_print helper

* sampling : hide prev behind API and apply #3661

ggml-ci
 2023-10-20 21:07:23 +03:00
Georgi Gerganov
778c070d1b
server : logs + minor code style 2023-10-20 20:44:51 +03:00
Georgi Gerganov
5d540e80d1
server : no need for atomic int - already using mutex 2023-10-20 20:44:29 +03:00
Georgi Gerganov
113dd60005
server : bach has to be allocated for n_parallel sequences 2023-10-20 20:42:45 +03:00
FSSRepo
6b2437e32d added thread safe pipeline 2023-10-20 12:07:32 -04:00
Qin Yue Chen
8cf19d60dc
gguf : support big endian platform (#3552) 
* check whether platform is 390x if yes->do not import immintrin.h

* support s390x big endian

* support --bigendian option for s390x
1. verified with baichuan7b-chat with float 16 on s390x
2. verified with baichuan7b-chat
3. verified with chinese-alpaca-2-13b-f16

* update format based on editor-config checker result

* Update convert-baichuan-hf-to-gguf.py

* 1. check in ggml.c if endianess is not match
2. update GGUF version
3. change get_pack_prefix to property
4. update information log

* always use "GGUF" as beginng of GGUF file

* Compare "GGUF" with file header char by char
1.  Set GGUF_MAGIC to "GGUF" string instead of int value
2. Compare "GGUF" char by char to ensure its byte order
3. Move bytes swap code from convert.py to gguf.py write_tensor_data

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
 2023-10-20 14:19:40 +03:00
Georgi Gerganov
a0edf73bda
server : fix uninitialized sampling context (close #3685) 2023-10-20 13:06:10 +03:00
Herman Semenov
f439e506e8
ggml : fix rope + llama minor optimizations (#3560) 
* Minor fixes and fixed memleak

* Using const auto references in range-based loop C++17
 2023-10-20 13:02:12 +03:00
cebtenzzre
e78f3ef24a
convert : restore compat with old Falcon models (#3680) 2023-10-20 08:32:08 +03:00
M. Yusuf Sarıgöz
f3b25e4043
multimodal : add BakLLaVA conversion support (#3682) 2023-10-19 19:40:41 +03:00
M. Yusuf Sarıgöz
60abea9798
llava : avoid segfault in case of non-existent mmproj file (#3674) 2023-10-19 16:59:11 +03:00
Georgi Gerganov
325d1793f7
server : minor sync 2023-10-19 15:03:24 +03:00
Georgi Gerganov
9740824ba5
server : snake case 2023-10-19 14:44:37 +03:00
Georgi Gerganov
e3a2c3fe32
server : use refs + use llama_batch_clear() 2023-10-19 14:44:04 +03:00
Georgi Gerganov
3d5929e8ee
server : bug fix in ingest_images 
n_tokens is incremented internally by llama_batch_add
 2023-10-19 14:43:19 +03:00
Georgi Gerganov
a8c981b734
server : remove beam-search functionality 2023-10-19 14:10:37 +03:00
Georgi Gerganov
654e0a1fe0
server : coding-style normalization (part 2) 2023-10-19 14:09:45 +03:00
Georgi Gerganov
e44ed60187
server : coding-style normalization 2023-10-19 13:50:23 +03:00
FSSRepo
ab2fc00224 latest changes of sampling API 2023-10-18 16:57:48 -04:00
FSSRepo
8540568c48 Merge branch 'master' of https://github.com/ggerganov/llama.cpp 2023-10-18 16:55:26 -04:00
FSSRepo
7196c4e08a new sampling API 2023-10-18 16:50:09 -04:00
Georgi Gerganov
004797f6ac
readme : update hot topics 2023-10-18 21:44:43 +03:00
Georgi Gerganov
4e82b2ea3f
speculative : bug fixes 2023-10-18 18:49:40 +03:00
Georgi Gerganov
0e89203b51
speculative : add tree-based sampling example (#3624) 
* sampling : one sequence per sampling context

ggml-ci

* speculative : add tree-based sampling support

ggml-ci

* speculative : reuse the n_parallel CLI param

* speculative : refactor sampling

* examples : fix build after sampling refactoring

ggml-ci

* batched : fix n_seq_id

* sampling : fix malloc

ggml-ci

* swift : fix build

ggml-ci

* swift : try to fix build

ggml-ci

* prompts : add assistant.txt

* common : add llama_batch_add() and llama_batch_clear() helpers

* speculative : minor refactor

ggml-ci

* minor : comments + rename

ggml-ci

* speculative : fix off-by-one for n_drafted

* speculative : fix the n_drafted fix + p constants
 2023-10-18 16:21:57 +03:00
Steward Garcia
84b8f2b060
Merge branch 'ggerganov:master' into master 2023-10-18 08:43:17 -04:00
Jhen-Jie Hong
c67fe68e41
metal : implement q5_0 and q5_1 kernels (#3648) 
* metal : implement dequantize_q5_0

* metal : block_q_n_dot_y for block_q5_0 (broken)

* metal : revert unnecessary change

* metal : implement dequantize_q5_1

* metal : block_q_n_dot_y for q5_1 (broken)

* metal : fix block_q_n_dot_y

* minor : spaces / formatting

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
 2023-10-18 15:21:48 +03:00
shibe2
1117d06607
opencl : fix element-wise multiplication (#3656) 2023-10-18 15:09:22 +03:00
FSSRepo
35fd37430f fix zig build 2023-10-17 18:04:26 -04:00
FSSRepo
c02c52efb5 fix multiple clients 2023-10-17 17:54:56 -04:00
FSSRepo
d2b1fac6c7 fix make bui;d errors 2023-10-17 17:18:56 -04:00
FSSRepo
ed0c11cb83 multimodal support enabled by default 2023-10-17 16:58:20 -04:00
FSSRepo
6c277eaab5 update api like OpenAI 2023-10-17 16:53:38 -04:00
FSSRepo
58f8ae9bfe readme change 2023-10-17 16:32:19 -04:00
FSSRepo
fa0f22f14f Merge remote-tracking branch 'upstream/master' 2023-10-17 16:31:33 -04:00
slaren
cb33f43a2a
fix embeddings when using CUDA (#3657) 2023-10-17 22:24:50 +02:00
FSSRepo
aa2268f4cd sync README.md changes 2023-10-17 16:21:05 -04:00
Georgi Gerganov
e1675d133c
llama : avoid fprintf in favor of LLAMA_LOG (#3538) 2023-10-17 22:34:26 +03:00
BarfingLemurs
8402566a7c
readme : update hot-topics & models, detail windows release in usage (#3615) 
* Update README.md

* Update README.md

* Update README.md

* move "Running on Windows" section below "Prepare data and run"

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
 2023-10-17 21:13:21 +03:00
shibe2
40e5ce054f CLBlast: Fix temporary buffer size for f16 conversion (wsize) 
Fix buffer overflow.
Reduce the size to fit just one 2D slice.
Assert sufficient size.
 2023-10-17 21:02:30 +04:00
slaren
a5e8c1d8c7
train-text-from-scratch : fix assert failure in ggml-alloc (#3618) 2023-10-17 20:00:58 +03:00
Georgi Gerganov
e74c705e15
editorconfig : remove trailing spaces 2023-10-17 19:52:53 +03:00
coezbek
3ad1e3f1a1
server : documentation of JSON return value of /completion endpoint (#3632) 
* Added documentation of JSON return value of /completion endpoint

* Update examples/server/README.md

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
 2023-10-17 19:51:02 +03:00