v0.4.18

Changed:

• llama.cpp revision eb7cf15a808d4d7a71eef89cc6a9b96fe82989dc

v0.4.17

Changed:

• llama.cpp revision 6152129d05870cb38162c422c6ba80434e021e9f

Fixed:

• Fixed build process, json patches.
• Reverted server code to previous version due to bug.

v0.3.1

Added:

• llama-cpp-cffi server - support for dynamic load/unload of model - hot-swap of models on demand
• llama-cpp-cffi server - compatible with llama.cpp cli options
• llama-cpp-cffi server - limited compatibility for OpenAI API /v1/chat/completions for text and vision models
• Support for CompletionsOptions.messages for VLM prompts with a single message containing just a pair of text and image_url in content.

Changed:

• llama.cpp revision 0827b2c1da299805288abbd556d869318f2b121e

v0.3.0

Added:

• Qwen 2 VL 2B / 7B vision models support
• WIP llama-cpp-cffi server - compatible with llama.cpp cli options instead of OpenAI

Changed:

• llama.cpp revision 5896c65232c7dc87d78426956b16f63fbf58dcf6
• Refactored Options class into two separate classes: ModelOptions, CompletionsOptions

Fixed:

• Llava (moondream2, nanoLLaVA-1.5, llava-v1.6-mistral-7b) vision models support
• MiniCPM-V 2.5 / 2.6 vision models support

Removed:

• Removed ambiguous Options class

v0.2.0

Added:

• New high-level Python API
• Low-level C API calls from llama.h, llava.h, clip.h, ggml.h
• completions for high-level function for LLMs / VLMs
• text_completions for low-level function for LLMs
• clip_completions for low-level function for CLIP-based VLMs
• WIP: mllama_completions for low-level function for Mllama-based VLMs

Changed:

• All examples

Removed:

• llama_generate function
• llama_cpp_cli
• llava_cpp_cli
• minicpmv_cpp_cli

v0.1.22

Added:

• llava high-level API calls
• minicpmv high-level API support

v0.1.16

Added:
- Updated llama.cpp.

v0.1.15

Added:
- SmolLM-1.7B-Instruct-v0.2 examples.
- Updated llama.cpp.

v0.1.14

Fixed:
- Vulkan detection.

0.1.13

Fixed:
- CUDA and Vulkan detection.