[workspace]

[package]
name = "pipeline-parallel-inference"
version = "0.1.0"
edition = "2024"
publish = false
[dependencies]
swactor = { path = "../..", features = ["transport", "serde", "std"] }
swactor-transport = { path = "../../crates/transport" }
# `ssh` enables the orchestrator's boot-phase telemetry: it SSHes into each
# rented node and streams the worker boot log onto its datastream.
swactor-process = { path = "../../crates/process", features = ["ssh"] }
serde = { version = "1", features = ["derive"] }
serde_json = "1"
reqwest = { version = "0.12", features = ["json", "stream"] }
futures-util = "0.3"
tokio = { version = "1", features = ["full"] }
distribution = { path = "../../crates/distribution" }
iroh-driver = { path = "../../crates/iroh-driver" }
datastream = { path = "../../crates/datastream" }
iroh = "0.98"
urlencoding = "2"
base64 = "0.22"
libc = "0.2"
swactor-vastai = { path = "../../tools/vastai" }

[[bin]]
name = "pp-worker"
path = "src/bin/pp_worker.rs"

[[bin]]
name = "pp-orchestrator"
path = "src/bin/pp_orchestrator.rs"

[dev-dependencies]
wiremock = "0.6"
tracing-subscriber = { version = "0.3", features = ["env-filter"] }

# These binaries are shipped to rented GPU nodes over the docker image on a
# cold lease, and over scp through vast.ai's throttled SSH proxy for manual
# swaps. Stripping debug symbols (~26MB -> ~18MB) trims both at no runtime cost.
[profile.release]
strip = true