v0.1.12

Added:
- Build vulkan_1_x for general GPU.
- Build cuda 12.4.1 as default.

Changed:
- Renamed examples for TinyLlama (chat, tool calling) and OpenAI.
- Updated demo models definitions.
- Updated examples (chat, tool calling).
- get_special_tokens not supports parameter force_standard_special_tokens: bool=False which bypasses tokenizer's special tokens with standard/common ones.
- Build cuda 12.5.1 as additional build target but packaged on PyPI.
- Build cuda 12.6 as additional build target but packaged on PyPI.
- Build openblas as additional build target but packaged on PyPI.

Fixed:
- Handle Options.no_display_prompt on Python side.

v0.1.11

Changed:
- openai: allow import of routes and v1_chat_completions handler.
- examples/demo_0.py, tool calling example

v0.1.10

Added:
- In openai, support for prompt and extra_body. Reference: https://github.com/openai/openai-python/blob/195c05a64d39c87b2dfdf1eca2d339597f1fce03/src/openai/resources/completions.py#L41
- Pass llama-cli options to openai.
- util module with is_cuda_available function.
- openai supports both prompt and messages. Reference: https://github.com/openai/openai-python/blob/195c05a64d39c87b2dfdf1eca2d339597f1fce03/src/openai/resources/completions.py#L45

v0.1.9

Added:
- Support for default CPU tinyBLAS (llamafile, sgemm) builds.
- Support for CPU OpenBLAS (GGML_OPENBLAS) builds.

Changed:
- Build scripts now have separate step/function cuda_12_5_1_setup which setups CUDA 12.5.1 env for build-time.

Fixed:
- Stop thread in llama_generate on GeneratorExit.

Removed:
- callback parameter in llama_generate and dependent functions.

v0.1.8

Added:
- Model.tokenizer_hf_repo as optional in case when Model.creator_hf_repo cannot be used to tokenize / format prompt/messages.

v0.1.7

Added:
- Support for stop tokens/words.

Changed:
- llama/llama_cli.py unified CPU and CUDA 12.5 modules into single module.

Removed:
- Removed separate examples for CPU and CUDA 12.5 modules.

v0.1.6

Changed:
- Updated huggingface-hub.

Fixed:
- llama.__init__ now correctly imports submodules and handles CPU and CUDA backends.
- OpenAI: ctx_size: int = config.max_position_embeddings if max_tokens is None else max_tokens.

v0.1.5

Fixed:
- Build for linux, upx uses best compression option, 7z uses more aggressive compression.
- Do not use UPX for shared/dynamic library compression.