[package]
edition = "2021"
rust-version = "1.85"
name = "vllm-cpp"
version = "0.0.1"
build = false
autolib = false
autobins = false
autoexamples = false
autotests = false
autobenches = false
description = "Safe model inference, streaming, chat, and concurrent requests for vllm.cpp"
documentation = "https://docs.rs/vllm-cpp"
readme = "README.md"
keywords = [
"llm",
"inference",
"ffi",
"vllm",
"machine-learning",
]
categories = [
"api-bindings",
"science",
]
license = "MIT OR Apache-2.0"
repository = "https://github.com/querymt/vllm-cpp-rs"
[package.metadata.docs.rs]
features = ["serde"]
targets = ["x86_64-unknown-linux-gnu"]
[features]
bundled = ["vllm-cpp-sys/bundled"]
cuda = ["vllm-cpp-sys/cuda"]
cuda-cutlass = [
"cuda",
"vllm-cpp-sys/cuda-cutlass",
]
default = ["bundled"]
dynamic-link = ["vllm-cpp-sys/dynamic-link"]
metal = ["vllm-cpp-sys/metal"]
mlx = [
"metal",
"vllm-cpp-sys/mlx",
]
serde = []
system = ["vllm-cpp-sys/system"]
triton-aot = [
"cuda",
"vllm-cpp-sys/triton-aot",
]
vulkan = ["vllm-cpp-sys/vulkan"]
[lib]
name = "vllm_cpp"
path = "src/lib.rs"
[[example]]
name = "chat"
path = "examples/chat.rs"
[[example]]
name = "complete"
path = "examples/complete.rs"
[[example]]
name = "concurrent"
path = "examples/concurrent.rs"
[[example]]
name = "setup_test_model"
path = "examples/setup_test_model.rs"
[[example]]
name = "stream"
path = "examples/stream.rs"
[[example]]
name = "structured"
path = "examples/structured.rs"
[[test]]
name = "qwen3"
path = "tests/qwen3.rs"
[[test]]
name = "safe_api"
path = "tests/safe_api.rs"
[dependencies.hf-hub]
version = "0.5.0"
features = ["ureq"]
default-features = false
[dependencies.serde_json]
version = "1.0.149"
[dependencies.vllm-cpp-sys]
version = "=0.0.1"
default-features = false
[dev-dependencies.clap]
version = "=4.6.1"
features = ["derive"]
[dev-dependencies.static_assertions]
version = "1.1.0"