[project]
name = "llm-inference"
version = "0.1.0"
description = "Unified LLM inference with multiple backends (LiteLLM, vLLM, llama.cpp)"
requires-python = ">=3.10"
readme = "README.md"

dependencies = [
    # Core dependencies (always installed)
    "litellm>=1.50.0,<2.0",
    "fastapi>=0.115.0,<0.116",
    "uvicorn[standard]>=0.32.0",
    "pydantic>=2.0,<3.0",
    "pydantic-settings>=2.0,<3.0",
    "httpx>=0.27.0,<1.0",
    "sse-starlette>=2.0,<3.0",
    "prometheus-client>=0.20.0",
    "prometheus-fastapi-instrumentator>=7.0.0",
    "opentelemetry-instrumentation-fastapi>=0.50b0",
    "opentelemetry-exporter-otlp-proto-grpc>=1.30.0",
]

[project.optional-dependencies]
# vLLM backend (GPU inference via OpenAI-compatible API)
vllm = [
    "openai>=1.50.0,<2.0",
]

# llama.cpp backend (CPU/Metal inference via OpenAI-compatible API)
llamacpp = [
    "openai>=1.50.0,<2.0",
]

# All local inference backends
local = [
    "llm-inference[vllm,llamacpp]",
]

# Development dependencies
dev = [
    "pytest>=8.0",
    "pytest-cov>=4.0",
    "pytest-asyncio>=0.24.0",
    "ruff>=0.8",
    "mypy>=1.0",
    "respx>=0.21.0",
]

[project.scripts]
llm-inference = "llm_inference.cli:main"

[build-system]
requires = ["hatchling"]
build-backend = "hatchling.build"

[tool.hatch.build.targets.wheel]
packages = ["src/llm_inference"]

[tool.ruff]
extend = "../../ruff.toml"

[tool.pytest.ini_options]
asyncio_mode = "auto"
testpaths = ["tests"]
addopts = "-v --tb=short"
markers = [
    "e2e: End-to-end tests against real running server",
    "slow: Tests that make real LLM API calls (cost money, slower)",
]

[tool.mypy]
python_version = "3.10"
strict = true
warn_return_any = true
warn_unused_ignores = true

[dependency-groups]
dev = [
    "pytest>=9.0.2",
    "pytest-asyncio>=1.3.0",
    "ruff>=0.14.11",
]