Commit graph

623 commits

Author SHA1 Message Date
Andrei
153a0049d9
feat: Generic chatml Function Calling (#957)
* Add demo notebook

* Add initial chat handler

* Update OpenAI types

* Add generic chatml function calling (wip)

* Update chatml generic function calling.

* Progress on auto-tool calls

* fix streaming functions

* Remove print statements

* fix: Suppress output from llama.cpp init and grammar creation

* Add OpenAI v1 python api compatible chat completion function

* Support non-streaming multi-tool calls

* Format

* Include function_call in response.
2024-02-12 15:56:07 -05:00
Andrei Betlen
69413ce08e Update llama.cpp 2024-02-11 19:00:17 -05:00
Connor
a05d90446f
fix: Circular dependancy preventing early Llama object free (#1176)
commit 901827013b introduced a cyclic dependency
within Llama objects. That change causes old models to linger in memory longer
than necessary, thereby creating memory bloat in most applications attempting
to switch between models at runtime. This patch simply removes the problematic
line, allowing models to deallocate without relying on GC. One might also
consider combining `weakref.ref` with a `@property` if the `llama` attribute is
absolutely necessary to expose in the tokenizer class.
2024-02-11 13:57:57 -05:00
Andrei Betlen
4abb8c9386 Merge branch 'main' of github.com:abetlen/llama_cpp_python into main 2024-02-09 13:32:31 -05:00
Andrei Betlen
e16f06e6eb fix: revert _create_completions. 2024-02-09 02:02:13 -05:00
Andrei Betlen
85d3374b4d fix: broken import 2024-02-08 01:13:28 -05:00
Andrei Betlen
b5fca911b5 feat: Move tokenizer to own module 2024-02-08 01:08:18 -05:00
Jeffrey Fong
901827013b
feat: Integrate functionary v1.4 and v2 models + add custom tokenizer support to Llama class (#1078)
* convert functionary-v1 chat handler to use hf autotokenizer

* add hf_tokenizer + inteegrate functionary-v1.4 prompt template

* integrate functionary v2 prompt template

* update readme

* set up parallel function calling wip

* set up parallel function calling

* Update README.md

* Update README.md

* refactor tokenizers

* include old functionary handler for backward compatibility

* add hf_tokenizer_path in server ModelSettings

* convert functionary-v1 chat handler to use hf autotokenizer

* add hf_tokenizer + inteegrate functionary-v1.4 prompt template

* integrate functionary v2 prompt template

* update readme

* set up parallel function calling wip

* resolve merge conflict

* Update README.md

* Update README.md

* refactor tokenizers

* include old functionary handler for backward compatibility

* add hf_tokenizer_path in server ModelSettings

* Cleanup PR, fix breaking changes

* Use hf_pretrained_model_name_or_path for tokenizer

* fix hf tokenizer in streaming

* update README

* refactor offset mapping

---------

Co-authored-by: Andrei <abetlen@gmail.com>
2024-02-07 20:07:03 -05:00
Andrei Betlen
34f31040f6 Bump version 2024-02-06 12:47:59 -05:00
Andrei Betlen
59760c85ed fix: Use llama_log_callback to avoid suppress_stdout_stderr 2024-02-05 21:52:12 -05:00
Andrei Betlen
3553b14670 Update llama.cpp 2024-02-05 13:26:50 -05:00
Andrei
7467f129e5
Revert "Fix: fileno error google colab (#729) (#1156)" (#1157)
This reverts commit bebfba0f08.
2024-02-02 12:18:55 -05:00
Dulsara
bebfba0f08
Fix: fileno error google colab (#729) (#1156)
Instead of using a devnull just made a dummy class with a 'write()' method that does nothing.
2024-02-02 12:05:46 -05:00
Andrei Betlen
3322eadbf3 Bump version 2024-01-31 15:10:18 -05:00
Andrei
fb762a6041
Add speculative decoding (#1120)
* Add draft model param to llama class, implement basic prompt lookup decoding draft model

* Use samplingcontext for sampling

* Use 1d array

* Use draft model for sampling

* Fix dumb mistake

* Allow for later extensions to the LlamaDraftModel api

* Cleanup

* Adaptive candidate prediction

* Update implementation to match hf transformers

* Tuning

* Fix bug where last token was not used for ngram prediction

* Remove heuristic for num_pred_tokens (no benefit)

* fix: n_candidates bug.

* Add draft_model_num_pred_tokens server setting

* Cleanup

* Update README
2024-01-31 14:08:14 -05:00
Andrei Betlen
71e3e4c435 Update llama.cpp 2024-01-31 10:41:42 -05:00
Andrei Betlen
078cca0361 fix: Pass raise_exception and add_generation_prompt to jinja2 chat template 2024-01-31 08:42:21 -05:00
Andrei Betlen
bf9e824922 Bump version 2024-01-30 12:27:27 -05:00
Andrei Betlen
011cd84ded Update llama.cpp 2024-01-30 09:48:09 -05:00
Andrei
da003d8768
Automatically set chat format from gguf (#1110)
* Use jinja formatter to load chat format from gguf

* Fix off-by-one error in metadata loader

* Implement chat format auto-detection
2024-01-29 14:22:23 -05:00
Andrei Betlen
464af5b39f Bump version 2024-01-29 10:46:04 -05:00
Andrei Betlen
9ae5819ee4 Add chat format test. 2024-01-29 00:59:01 -05:00
Rafaelblsilva
ce38dbdf07
Add mistral instruct chat format as "mistral-instruct" (#799)
* Added mistral instruct chat format as "mistral"

* Fix stop sequence (merge issue)

* Update chat format name to `mistral-instruct`

---------

Co-authored-by: Andrei <abetlen@gmail.com>
2024-01-29 00:34:42 -05:00
Andrei Betlen
52c4a84faf Bump version 2024-01-28 19:35:37 -05:00
Andrei Betlen
c1d0fff8a9 Bump version 2024-01-27 18:36:56 -05:00
Andrei
d8f6914f45
Add json schema mode (#1122)
* Add json schema mode

* Add llava chat format support
2024-01-27 16:52:18 -05:00
Andrei Betlen
35918873b4 Update llama.cpp 2024-01-26 11:45:48 -05:00
Andrei Betlen
f5cc6b3053 Bump version 2024-01-25 11:28:16 -05:00
Andrei Betlen
cde7514c3d feat(server): include llama-cpp-python version in openapi spec 2024-01-25 11:23:18 -05:00
Andrei Betlen
c970d41a85 fix: llama_log_set should be able to accept null pointer 2024-01-24 10:38:30 -05:00
Andrei Betlen
9677a1f2c8 fix: Check order 2024-01-23 22:28:03 -05:00
Andrei Betlen
4d6b2f7b91 fix: format 2024-01-23 22:08:27 -05:00
Phil H
fe5d6ea648
fix: GGUF metadata KV overrides, re #1011 (#1116)
* kv overrides another attempt

* add sentinel element, simplify array population

* ensure sentinel element is zeroed
2024-01-23 22:00:38 -05:00
Andrei Betlen
fcdf337d84 Update llama.cpp 2024-01-22 11:25:11 -05:00
Andrei Betlen
5b982d0f8c fix: use both eos and bos tokens as stop sequences for hf-tokenizer-config chat format. 2024-01-22 08:32:48 -05:00
Andrei Betlen
2ce0b8aa2c Bump version 2024-01-21 20:30:24 -05:00
Andrei Betlen
d3f5528ca8 fix: from_json_schema oneof/anyof bug. Closes #1097 2024-01-21 19:06:53 -05:00
Andrei Betlen
24f39454e9 fix: pass chat handler not chat formatter for huggingface autotokenizer and tokenizer_config formats. 2024-01-21 18:38:04 -05:00
Andrei Betlen
7f3209b1eb feat: Add add_generation_prompt option for jinja2chatformatter. 2024-01-21 18:37:24 -05:00
Andrei Betlen
be09318c26 feat: Add Jinja2ChatFormatter 2024-01-19 15:04:42 -05:00
Andrei Betlen
5a34c57e54 feat: Expose gguf model metadata in metadata property 2024-01-19 10:46:03 -05:00
Andrei Betlen
833a7f1a86 Bump version 2024-01-19 09:03:35 -05:00
Andrei Betlen
3babe3512c Fix mirostat sampling 2024-01-19 08:31:59 -05:00
Andrei Betlen
141293a75b Fix python3.8 support 2024-01-19 08:17:49 -05:00
Andrei Betlen
656f3d8968 Bump version 2024-01-18 21:30:36 -05:00
Andrei Betlen
89cce50f8c Update llama.cpp 2024-01-18 21:21:49 -05:00
Andrei Betlen
b8fc1c7d83 feat: Add ability to load chat format from huggingface autotokenizer or tokenizer_config.json files. 2024-01-18 21:21:37 -05:00
Andrei Betlen
48c3b77e6f Offload KQV by default 2024-01-18 11:08:57 -05:00
Austin
6bfe98bd80
Integration of Jinja2 Templating (#875)
* feat: Add support for jinja templating

Signed-off-by: teleprint-me <77757836+teleprint-me@users.noreply.github.com>

* fix: Refactor chat formatter and update interface for jinja templates

- Simplify the `llama2_template` in `llama_jinja_format.py` by removing unnecessary line breaks for readability without affecting functionality.
- Update `ChatFormatterInterface` constructor to accept a more generic `Optional[object]` type for the template parameter, enhancing flexibility.
- Introduce a `template` property to `ChatFormatterInterface` for standardized access to the template string.
- Replace `MetaSingleton` metaclass with `Singleton` for the `ChatFormatterFactory` to streamline the singleton implementation.

These changes enhance code readability, maintain usability, and ensure consistency in the chat formatter's design pattern usage.

* Add outline for Jinja2 templating integration documentation

Signed-off-by: teleprint-me <77757836+teleprint-me@users.noreply.github.com>

* Add jinja2 as a dependency with version range for Hugging Face transformers compatibility

Signed-off-by: teleprint-me <77757836+teleprint-me@users.noreply.github.com>

* Update jinja2 version constraint for mkdocs-material compatibility

Signed-off-by: teleprint-me <77757836+teleprint-me@users.noreply.github.com>

* Fix attribute name in AutoChatFormatter

- Changed attribute name from `self._renderer` to `self._environment`

---------

Signed-off-by: teleprint-me <77757836+teleprint-me@users.noreply.github.com>
2024-01-17 09:47:52 -05:00
Andrei Betlen
7b46bb5a78 Re-order classes in llama.py 2024-01-17 09:16:13 -05:00