llama_cpp 0.17.0 → 0.17.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: a63238d7d4a852e4a57667ba3e144364db201a691b9460c62fc8aa783677593d
-  data.tar.gz: 7a879c04eebc5a308ae3f937f35972b11c5d15edd5000885416e3c57cfe21648
+  metadata.gz: 2b2fec35458bc9b745aa4e2526b2c50ca52201e8f29f608d84993b1eddff5a2f
+  data.tar.gz: 00f8d95bec17dcb422eb833623d5eca5028598e2a212dd71a248ad5f63434165
 SHA512:
-  metadata.gz: a76006fc44d8a7b4295c4d10bcee87a2f161868b9c119ddfae1c2aecd0a5d7989bd33134dc64d8f1994b41732a64e2ca91472a8245ee58e3fb4fdcb01a1b24f2
-  data.tar.gz: 63160f285f7fdb89e6d03e9cb83b064acbe8869ae384f9b3d32f0a822d7fc63354cf0fb6b6da39758140d885493baff716d31c42a956e3437c47adaf74172783
+  metadata.gz: fd7e98833df714d4c355820995e79964b74f31e0a4dc516360191a9c8c290108a5bf3d90b1ae704f1920ebb3db0152c2de17e2a8ec955fdc3ae1e979abae66ae
+  data.tar.gz: c1e32582670b1069187a1c2f8277296b0878c5dc613dca3c733378689086dc10fca2e1ee7d8e6aec6d9db95ebdc2bdfbb5b5c108808b5e489735d1ba19c52cd4

data/CHANGELOG.md

@@ -1,3 +1,9 @@
+## [[0.17.1](https://github.com/yoshoku/llama_cpp.rb/compare/v0.17.0...v0.17.1)] - 2024-07-06
+
+- Update usage section on README.
+- Change supported llama.cpp version to b3291.
+  - Add `LLAMA_VOCAB_PRE_TYPE_JAIS` constant.
+
 ## [[0.17.0](https://github.com/yoshoku/llama_cpp.rb/compare/v0.16.2...v0.17.0)] - 2024-06-29
 
 **Breaking Changes**
@@ -12,7 +18,7 @@ $ brew install llama.cpp
 $ gem install llama_cpp -- --with-opt-dir=/opt/homebrew
 ```
 
-- Change supported llama.cpp version to b3265
+- Change supported llama.cpp version to b3265.
  - Add `LLAMA_VOCAB_TYPE_UGM` and `LLAMA_VOCAB_PRE_TYPE_VIKING` constants.
  - Add `token_pad` method to `Model`.
 

data/README.md

@@ -32,7 +32,10 @@ $ gem install llama_cpp -- --with-opt-dir=/opt/homebrew
 ## Usage
 
 Prepare the quantized model by refering to [the usage section on the llama.cpp README](https://github.com/ggerganov/llama.cpp#usage).
-For example, preparing the quatization model based on [open_llama_7b](https://huggingface.co/openlm-research/open_llama_7b) is as follows:
+For example, you could prepare the quatization model based on
+[open_llama_7b](https://huggingface.co/openlm-research/open_llama_7b)
+or more useful in the context of Ruby might be a smaller model such as
+[tiny_llama_1b](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0):
 
 ```sh
 $ cd ~/
@@ -44,9 +47,9 @@ $ python3 -m pip install -r requirements.txt
 $ cd models
 $ git clone https://huggingface.co/openlm-research/open_llama_7b
 $ cd ../
-$ python3 convert.py models/open_llama_7b
+$ python3 convert-hf-to-gguf.py models/open_llama_7b
 $ make
-$ ./quantize ./models/open_llama_7b/ggml-model-f16.gguf ./models/open_llama_7b/ggml-model-q4_0.bin q4_0
+$ ./llama-quantize ./models/open_llama_7b/ggml-model-f16.gguf ./models/open_llama_7b/ggml-model-q4_0.bin q4_0
 ```
 
 An example of Ruby code that generates sentences with the quantization model is as follows:

data/ext/llama_cpp/llama_cpp.cpp

@@ -3514,6 +3514,7 @@ extern "C" void Init_llama_cpp(void) {
   rb_define_const(rb_mLLaMACpp, "LLAMA_VOCAB_PRE_TYPE_SMAUG", INT2NUM(LLAMA_VOCAB_PRE_TYPE_SMAUG));
   rb_define_const(rb_mLLaMACpp, "LLAMA_VOCAB_PRE_TYPE_PORO", INT2NUM(LLAMA_VOCAB_PRE_TYPE_PORO));
   rb_define_const(rb_mLLaMACpp, "LLAMA_VOCAB_PRE_TYPE_VIKING", INT2NUM(LLAMA_VOCAB_PRE_TYPE_VIKING));
+  rb_define_const(rb_mLLaMACpp, "LLAMA_VOCAB_PRE_TYPE_JAIS", INT2NUM(LLAMA_VOCAB_PRE_TYPE_JAIS));
 
   rb_define_const(rb_mLLaMACpp, "LLAMA_TOKEN_TYPE_UNDEFINED", INT2NUM(LLAMA_TOKEN_TYPE_UNDEFINED));
   rb_define_const(rb_mLLaMACpp, "LLAMA_TOKEN_TYPE_NORMAL", INT2NUM(LLAMA_TOKEN_TYPE_NORMAL));

data/lib/llama_cpp/version.rb

@@ -3,8 +3,8 @@
 # llama_cpp.rb provides Ruby bindings for the llama.cpp.
 module LLaMACpp
   # The version of llama_cpp.rb you install.
-  VERSION = '0.17.0'
+  VERSION = '0.17.1'
 
   # The supported version of llama.cpp.
-  LLAMA_CPP_VERSION = 'b3265'
+  LLAMA_CPP_VERSION = 'b3291'
 end

data/sig/llama_cpp.rbs

@@ -34,6 +34,7 @@ module LLaMACpp
   LLAMA_VOCAB_PRE_TYPE_SMAUG: Integer
   LLAMA_VOCAB_PRE_TYPE_PORO: Integer
   LLAMA_VOCAB_PRE_TYPE_VIKING: Integer
+  LLAMA_VOCAB_PRE_TYPE_JAIS: Integer
 
   LLAMA_TOKEN_ATTR_UNDEFINED: Integer
   LLAMA_TOKEN_ATTR_UNKNOWN: Integer

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: llama_cpp
 version: !ruby/object:Gem::Version
-  version: 0.17.0
+  version: 0.17.1
 platform: ruby
 authors:
 - yoshoku
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2024-06-29 00:00:00.000000000 Z
+date: 2024-07-06 00:00:00.000000000 Z
 dependencies: []
 description: llama_cpp.rb provides Ruby bindings for the llama.cpp.
 email: