llama_cpp 0.17.0 → 0.17.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/CHANGELOG.md +7 -1
- data/README.md +6 -3
- data/ext/llama_cpp/llama_cpp.cpp +1 -0
- data/lib/llama_cpp/version.rb +2 -2
- data/sig/llama_cpp.rbs +1 -0
- metadata +2 -2
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 2b2fec35458bc9b745aa4e2526b2c50ca52201e8f29f608d84993b1eddff5a2f
|
|
4
|
+
data.tar.gz: 00f8d95bec17dcb422eb833623d5eca5028598e2a212dd71a248ad5f63434165
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: fd7e98833df714d4c355820995e79964b74f31e0a4dc516360191a9c8c290108a5bf3d90b1ae704f1920ebb3db0152c2de17e2a8ec955fdc3ae1e979abae66ae
|
|
7
|
+
data.tar.gz: c1e32582670b1069187a1c2f8277296b0878c5dc613dca3c733378689086dc10fca2e1ee7d8e6aec6d9db95ebdc2bdfbb5b5c108808b5e489735d1ba19c52cd4
|
data/CHANGELOG.md
CHANGED
|
@@ -1,3 +1,9 @@
|
|
|
1
|
+
## [[0.17.1](https://github.com/yoshoku/llama_cpp.rb/compare/v0.17.0...v0.17.1)] - 2024-07-06
|
|
2
|
+
|
|
3
|
+
- Update usage section on README.
|
|
4
|
+
- Change supported llama.cpp version to b3291.
|
|
5
|
+
- Add `LLAMA_VOCAB_PRE_TYPE_JAIS` constant.
|
|
6
|
+
|
|
1
7
|
## [[0.17.0](https://github.com/yoshoku/llama_cpp.rb/compare/v0.16.2...v0.17.0)] - 2024-06-29
|
|
2
8
|
|
|
3
9
|
**Breaking Changes**
|
|
@@ -12,7 +18,7 @@ $ brew install llama.cpp
|
|
|
12
18
|
$ gem install llama_cpp -- --with-opt-dir=/opt/homebrew
|
|
13
19
|
```
|
|
14
20
|
|
|
15
|
-
- Change supported llama.cpp version to b3265
|
|
21
|
+
- Change supported llama.cpp version to b3265.
|
|
16
22
|
- Add `LLAMA_VOCAB_TYPE_UGM` and `LLAMA_VOCAB_PRE_TYPE_VIKING` constants.
|
|
17
23
|
- Add `token_pad` method to `Model`.
|
|
18
24
|
|
data/README.md
CHANGED
|
@@ -32,7 +32,10 @@ $ gem install llama_cpp -- --with-opt-dir=/opt/homebrew
|
|
|
32
32
|
## Usage
|
|
33
33
|
|
|
34
34
|
Prepare the quantized model by refering to [the usage section on the llama.cpp README](https://github.com/ggerganov/llama.cpp#usage).
|
|
35
|
-
For example,
|
|
35
|
+
For example, you could prepare the quatization model based on
|
|
36
|
+
[open_llama_7b](https://huggingface.co/openlm-research/open_llama_7b)
|
|
37
|
+
or more useful in the context of Ruby might be a smaller model such as
|
|
38
|
+
[tiny_llama_1b](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0):
|
|
36
39
|
|
|
37
40
|
```sh
|
|
38
41
|
$ cd ~/
|
|
@@ -44,9 +47,9 @@ $ python3 -m pip install -r requirements.txt
|
|
|
44
47
|
$ cd models
|
|
45
48
|
$ git clone https://huggingface.co/openlm-research/open_llama_7b
|
|
46
49
|
$ cd ../
|
|
47
|
-
$ python3 convert.py models/open_llama_7b
|
|
50
|
+
$ python3 convert-hf-to-gguf.py models/open_llama_7b
|
|
48
51
|
$ make
|
|
49
|
-
$ ./quantize ./models/open_llama_7b/ggml-model-f16.gguf ./models/open_llama_7b/ggml-model-q4_0.bin q4_0
|
|
52
|
+
$ ./llama-quantize ./models/open_llama_7b/ggml-model-f16.gguf ./models/open_llama_7b/ggml-model-q4_0.bin q4_0
|
|
50
53
|
```
|
|
51
54
|
|
|
52
55
|
An example of Ruby code that generates sentences with the quantization model is as follows:
|
data/ext/llama_cpp/llama_cpp.cpp
CHANGED
|
@@ -3514,6 +3514,7 @@ extern "C" void Init_llama_cpp(void) {
|
|
|
3514
3514
|
rb_define_const(rb_mLLaMACpp, "LLAMA_VOCAB_PRE_TYPE_SMAUG", INT2NUM(LLAMA_VOCAB_PRE_TYPE_SMAUG));
|
|
3515
3515
|
rb_define_const(rb_mLLaMACpp, "LLAMA_VOCAB_PRE_TYPE_PORO", INT2NUM(LLAMA_VOCAB_PRE_TYPE_PORO));
|
|
3516
3516
|
rb_define_const(rb_mLLaMACpp, "LLAMA_VOCAB_PRE_TYPE_VIKING", INT2NUM(LLAMA_VOCAB_PRE_TYPE_VIKING));
|
|
3517
|
+
rb_define_const(rb_mLLaMACpp, "LLAMA_VOCAB_PRE_TYPE_JAIS", INT2NUM(LLAMA_VOCAB_PRE_TYPE_JAIS));
|
|
3517
3518
|
|
|
3518
3519
|
rb_define_const(rb_mLLaMACpp, "LLAMA_TOKEN_TYPE_UNDEFINED", INT2NUM(LLAMA_TOKEN_TYPE_UNDEFINED));
|
|
3519
3520
|
rb_define_const(rb_mLLaMACpp, "LLAMA_TOKEN_TYPE_NORMAL", INT2NUM(LLAMA_TOKEN_TYPE_NORMAL));
|
data/lib/llama_cpp/version.rb
CHANGED
|
@@ -3,8 +3,8 @@
|
|
|
3
3
|
# llama_cpp.rb provides Ruby bindings for the llama.cpp.
|
|
4
4
|
module LLaMACpp
|
|
5
5
|
# The version of llama_cpp.rb you install.
|
|
6
|
-
VERSION = '0.17.
|
|
6
|
+
VERSION = '0.17.1'
|
|
7
7
|
|
|
8
8
|
# The supported version of llama.cpp.
|
|
9
|
-
LLAMA_CPP_VERSION = '
|
|
9
|
+
LLAMA_CPP_VERSION = 'b3291'
|
|
10
10
|
end
|
data/sig/llama_cpp.rbs
CHANGED
metadata
CHANGED
|
@@ -1,14 +1,14 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: llama_cpp
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.17.
|
|
4
|
+
version: 0.17.1
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- yoshoku
|
|
8
8
|
autorequire:
|
|
9
9
|
bindir: exe
|
|
10
10
|
cert_chain: []
|
|
11
|
-
date: 2024-06
|
|
11
|
+
date: 2024-07-06 00:00:00.000000000 Z
|
|
12
12
|
dependencies: []
|
|
13
13
|
description: llama_cpp.rb provides Ruby bindings for the llama.cpp.
|
|
14
14
|
email:
|