add root dockerfile that contains binaries for all services

This commit is contained in:
geoffsee
2025-09-04 14:54:20 -04:00
parent fb5098eba6
commit 296d4dbe7e
13 changed files with 189 additions and 255 deletions

View File

@@ -1,7 +1,7 @@
[package]
name = "inference-engine"
version.workspace = true
edition = "2021"
edition = "2024"
[dependencies]
candle-core = { git = "https://github.com/huggingface/candle.git" }
@@ -31,14 +31,21 @@ utoipa = { version = "4.2.0", features = ["axum_extras"] }
uuid = { version = "1.7.0", features = ["v4"] }
reborrow = "0.5.5"
futures-util = "0.3.31"
gemma-runner = { path = "../../integration/gemma-runner", features = ["metal"] }
llama-runner = { path = "../../integration/llama-runner", features = ["metal"]}
gemma-runner = { path = "../../integration/gemma-runner" }
llama-runner = { path = "../../integration/llama-runner" }
embeddings-engine = { path = "../embeddings-engine" }
[target.'cfg(target_os = "linux")'.dependencies]
candle-core = { git = "https://github.com/huggingface/candle.git", default-features = false }
candle-nn = { git = "https://github.com/huggingface/candle.git", default-features = false }
candle-transformers = { git = "https://github.com/huggingface/candle.git", default-features = false }
[target.'cfg(target_os = "macos")'.dependencies]
candle-core = { git = "https://github.com/huggingface/candle.git", features = ["metal"] }
candle-nn = { git = "https://github.com/huggingface/candle.git", features = ["metal"] }
candle-transformers = { git = "https://github.com/huggingface/candle.git", features = ["metal"] }
gemma-runner = { path = "../../integration/gemma-runner", features = ["metal"] }
llama-runner = { path = "../../integration/llama-runner", features = ["metal"] }
[dev-dependencies]
@@ -62,15 +69,19 @@ bindgen_cuda = { version = "0.1.1", optional = true }
[features]
bin = []
[[bin]]
name = "inference-engine"
path = "src/main.rs"
[package.metadata.compose]
image = "ghcr.io/geoffsee/inference-engine:latest"
image = "ghcr.io/geoffsee/predict-otron-9000:latest"
cmd = ["./bin/inference-engine"]
port = 8080
# generates kubernetes manifests
[package.metadata.kube]
image = "ghcr.io/geoffsee/inference-service:latest"
replicas = 1
image = "ghcr.io/geoffsee/predict-otron-9000:latest"
cmd = ["./bin/inference-engine"]
port = 8080
replicas = 1