Commit graph

1167 commits

Author SHA1 Message Date
Andrei Betlen
9ae9c86be0 Update server docs 2023-11-08 00:52:13 -05:00
Andrei Betlen
598780fde8 Update Multimodal notebook 2023-11-08 00:48:25 -05:00
Andrei Betlen
b30b9c338b Add JSON mode support. Closes #881 2023-11-08 00:07:16 -05:00
Andrei Betlen
4852a6a39c Fix built in GBNF grammar rules 2023-11-08 00:06:22 -05:00
Andrei Betlen
64f5153c35 Add seed parameter to chat handlers 2023-11-07 23:41:29 -05:00
Andrei Betlen
86aeb9f3a1 Add seed parameter support for completion and chat_completion requests. Closes #884 2023-11-07 23:37:28 -05:00
Andrei Betlen
da1b80285a Update changelog 2023-11-07 23:15:26 -05:00
Andrei Betlen
9a8e64d29d Update llama.cpp 2023-11-07 23:14:19 -05:00
Andrei Betlen
3660230faa Fix docs multi-modal docs 2023-11-07 22:52:08 -05:00
Damian Stewart
aab74f0b2b
Multimodal Support (Llava 1.5) (#821)
* llava v1.5 integration

* Point llama.cpp to fork

* Add llava shared library target

* Fix type

* Update llama.cpp

* Add llava api

* Revert changes to llama and llama_cpp

* Update llava example

* Add types for new gpt-4-vision-preview api

* Fix typo

* Update llama.cpp

* Update llama_types to match OpenAI v1 API

* Update ChatCompletionFunction type

* Reorder request parameters

* More API type fixes

* Even More Type Updates

* Add parameter for custom chat_handler to Llama class

* Fix circular import

* Convert to absolute imports

* Fix

* Fix pydantic Jsontype bug

* Accept list of prompt tokens in create_completion

* Add llava1.5 chat handler

* Add Multimodal notebook

* Clean up examples

* Add server docs

---------

Co-authored-by: Andrei Betlen <abetlen@gmail.com>
2023-11-07 22:48:51 -05:00
Andrei Betlen
56171cf7bf Bump version 2023-11-06 09:37:55 -05:00
Andrei Betlen
52320c348c Add python 3.12 classifier 2023-11-06 09:34:07 -05:00
Andrei Betlen
4286830f16 Add python3.12 tests 2023-11-06 09:32:20 -05:00
Andrei Betlen
be0add1b2d Fix type bug 2023-11-06 09:30:38 -05:00
Andrei Betlen
e214a58422 Refactor Llama class internals 2023-11-06 09:16:36 -05:00
Andrei Betlen
bbffdaebaa Refactor autotokenizer format to reusable function 2023-11-06 09:07:27 -05:00
Andrei Betlen
b0e597e46e Pin python version in release 2023-11-06 08:56:41 -05:00
Joe
4ff8def4d0
#717: Add support for Huggingface Autotokenizer (#790)
Co-authored-by: Andrei <abetlen@gmail.com>
2023-11-05 18:06:36 -05:00
earonesty
3580e2c5df
Update llama_chat_format.py (#869)
* Update llama_chat_format.py

properly formal llama2 with first-message prompt embedded

* Update llama_chat_format.py
2023-11-05 17:00:13 -05:00
Andrei Betlen
f0b30ef7dc Update llama.cpp 2023-11-05 16:57:10 -05:00
Andrei Betlen
dccbac82eb Update llama.cpp 2023-11-03 18:12:22 -04:00
Andrei Betlen
2ec043af76 Clean up stdout / stderr suppression 2023-11-03 13:02:15 -04:00
Andrei Betlen
4ea7027c41 Rename internal only module utils to _utils 2023-11-03 12:55:55 -04:00
Andrei Betlen
df9362eeea Update llama.cpp 2023-11-03 11:34:50 -04:00
Andrei
3af7b21ff1
Add functionary support (#784)
* Add common grammars and json-schema-to-grammar utility function from llama.cpp

* Pass functions to format function

* Add basic functionary formatting

* Add LlamaChatHandler for more complex chat use cases

* Add function calling example notebook

* Add support for regular chat completions alongside function calling
2023-11-03 02:12:14 -04:00
Andrei Betlen
df31303a12 Update CHANGELOG 2023-11-02 20:16:32 -04:00
Andrei
ab028cb878
Migrate inference to llama_batch and llama_decode api (#795)
* Add low-level batching notebook

* fix: tokenization of special characters: (#850)

It should behave like llama.cpp, where most out of the box usages
treat special characters accordingly

* Update CHANGELOG

* Cleanup

* Fix runner label

* Update notebook

* Use llama_decode and batch api

* Support logits_all parameter

---------

Co-authored-by: Antoine Lizee <antoine.lizee@gmail.com>
2023-11-02 20:13:57 -04:00
Andrei Betlen
f436e0c872 Update llama.cpp 2023-11-02 17:34:01 -04:00
Andrei Betlen
8350de9a18 Bump version 2023-11-02 15:53:01 -04:00
Andrei Betlen
9ffe62d665 Update llama.cpp 2023-11-02 15:45:27 -04:00
Andrei Betlen
011b95d7f3 Fix name 'open' is not defined exception. Closes #860 2023-11-02 15:30:55 -04:00
Andrei Betlen
fa83cc5f9c Update llama.cpp
Fix build examples

Exclude examples directory

Revert cmake changes

Try actions/checkout@v4

Try to update submodules

Revert

Update llama.cpp

Fix build examples

Exclude examples directory

Revert cmake changes

Try actions/checkout@v4

Try to update submodules

Revert
2023-11-02 14:28:15 -04:00
Andrei Betlen
ddbd10c442 Fix clblast test 2023-11-02 14:28:15 -04:00
Andrei Betlen
735522272b Fix runner label 2023-11-02 14:28:15 -04:00
Andrei Betlen
0feffb9c20 Cleanup 2023-11-02 14:28:15 -04:00
Andrei Betlen
7fe0bd3a31 Update CHANGELOG 2023-11-02 14:28:15 -04:00
Antoine Lizee
4d4e0f11e2 fix: tokenization of special characters: (#850)
It should behave like llama.cpp, where most out of the box usages
treat special characters accordingly
2023-11-02 14:28:14 -04:00
Andrei Betlen
952e4cc3ce Fix: use linux image for opencl test 2023-11-01 21:31:02 -04:00
Andrei Betlen
8bf7fa6e5f Add opencl test 2023-11-01 21:18:36 -04:00
Andrei Betlen
446d5f5649 Add metal ci test 2023-11-01 21:15:01 -04:00
Andrei Betlen
c89eadafbf Update CHANGELOG 2023-11-01 19:40:04 -04:00
Andrei Betlen
6b3aa7fc8f Bump version 2023-11-01 19:25:03 -04:00
NickAlgra
3fbcded7cd
Add missing n_seq_id to llama_batch (#842) 2023-11-01 18:56:29 -04:00
Sujeendran Menon
7b136bb5b1
Fix for shared library not found and compile issues in Windows (#848)
* fix windows library dll name issue

* Updated README.md Windows instructions

* Update llama_cpp.py to handle different windows dll file versions
2023-11-01 18:55:57 -04:00
cebtenzzre
eefd76fe81
llama: fix exception in Llama.__del__ (#846) 2023-11-01 18:53:57 -04:00
David Ponce
3fc9147218
Iterate over tokens that should be biased rather than the entire vocabulary. (#851) 2023-11-01 18:53:47 -04:00
Marko Tasic
9c8f4dca5f
fixed Llama._create_completion suffix check, it can be either None or str instance (#854) 2023-11-01 18:52:50 -04:00
Daniel Thuerck
5f8f369d1b
Pass-Through grammar parameter in web server. (#855) Closes #778 2023-11-01 18:51:12 -04:00
Adam Katora
25cb710281
Update llama_types.py (#849)
Minor typo fix, funcion -> function
2023-11-01 18:50:11 -04:00
Andrei Betlen
bdf5254658 Update llama.cpp 2023-11-01 14:15:56 -04:00