feat: add LiteLLM as AI gateway provider

docs/docs/tutorials/items/model_integration/llm_integration.ipynb

@@ -17,7 +17,8 @@
         "|:-------------------------:|:---------:|:------------------:|:---------------:|\n",
         "| `bridgic-llms-openai-like`|     ✅     |         ❌          |        ❌        |\n",
         "| `bridgic-llms-openai`     |     ✅     |         ✅          |        ✅        |\n",
-        "| `bridgic-llms-vllm`       |     ✅     |         ✅          |        ✅        |\n"
+        "| `bridgic-llms-vllm`       |     ✅     |         ✅          |        ✅        |\n",
+        "| `bridgic-llms-litellm`    |     ✅     |         ❌          |        ❌        |\n"
       ]
     },
     {

packages/bridgic-integration/llms/bridgic-llms-litellm/Makefile

@@ -0,0 +1,36 @@
+.PHONY: venv-init test build publish
+
+package_name := $(notdir $(CURDIR))
+repo ?= btsk
+
+ROOT_DIR := $(shell git rev-parse --show-toplevel)
+VERSION_CHECK := $(ROOT_DIR)/scripts/version_check.py
+SET_CREDENTIALS := $(ROOT_DIR)/scripts/set_publish_credentials.sh
+
+venv-collect:
+	@echo "\n==> Installing dependencies for [${package_name}]..."
+	@uv pip install -e .
+
+test:
+	@uv run -- pytest
+
+build:
+	@mkdir -p dist
+	@rm -rf dist/*
+	@package_name=$$(uv run python -c "import tomli; print(tomli.load(open('pyproject.toml', 'rb'))['project']['name'])") && \
+	uv build --package "$$package_name" --out-dir dist
+
+publish:
+	@source $(SET_CREDENTIALS) && \
+	version=$$(uv run python -c "import tomli; print(tomli.load(open('pyproject.toml', 'rb'))['project']['version'])") && \
+	uv run python $(VERSION_CHECK) --version "$$version" --repo "$(repo)" --package "$(package_name)" && \
+	$(MAKE) _publish_$(repo)
+
+_publish_btsk:
+	@uv publish dist/* --index btsk-repo --config-file $(ROOT_DIR)/uv.toml
+
+_publish_testpypi:
+	@uv publish dist/* --index test-pypi --config-file $(ROOT_DIR)/uv.toml
+
+_publish_pypi:
+	@uv publish dist/* --config-file $(ROOT_DIR)/uv.toml

packages/bridgic-integration/llms/bridgic-llms-litellm/README.md

@@ -0,0 +1,27 @@
+# bridgic-llms-litellm
+
+LiteLLM adapters for [Bridgic](https://github.com/bitsky-tech/bridgic), enabling connectivity with 100+ LLM providers through a single unified interface.
+
+## Installation
+
+```shell
+pip install bridgic-llms-litellm
+```
+
+## Usage
+
+```python
+from bridgic.llms.litellm import LiteLLM
+from bridgic.core.model.types import Message, Role
+
+# API keys are read from environment variables (e.g. OPENAI_API_KEY)
+llm = LiteLLM()
+
+response = llm.chat(
+    model="openai/gpt-4o",
+    messages=[Message.from_text("Hello!", role=Role.USER)],
+)
+print(response.message.content)
+```
+
+See https://docs.litellm.ai/docs/providers for the full provider list.

packages/bridgic-integration/llms/bridgic-llms-litellm/bridgic/llms/litellm/__init__.py

@@ -0,0 +1,22 @@
+"""
+The LiteLLM integration module provides access to 100+ LLM providers through
+a single unified interface.
+
+Supported providers include OpenAI, Anthropic, Google, Groq, Together AI,
+AWS Bedrock, Azure, Mistral, and many more.  Uses provider-prefixed model
+names, e.g. ``openai/gpt-4o``, ``anthropic/claude-sonnet-4-6``.
+
+See https://docs.litellm.ai/docs/providers for the full provider list.
+
+You can install the LiteLLM integration package for Bridgic by running:
+
+```shell
+pip install bridgic-llms-litellm
+```
+"""
+
+from importlib.metadata import version
+from ._litellm_llm import LiteLLMConfiguration, LiteLLM
+
+__version__ = version("bridgic-llms-litellm")
+__all__ = ["LiteLLMConfiguration", "LiteLLM", "__version__"]