furnace/Cargo.toml at main · Gilfeather/furnace · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
[package]
name = "furnace"
version = "0.3.5"
edition = "2021"
rust-version = "1.75"
authors = ["Furnace Contributors"]
description = "🔥 High-performance Rust ML inference server powered by Burn framework"
documentation = "https://docs.rs/furnace"
homepage = "https://github.com/Gilfeather/furnace"
repository = "https://github.com/Gilfeather/furnace"
license = "MIT"
keywords = ["machine-learning", "inference", "server", "rust", "burn"]
categories = ["web-programming::http-server", "science"]
readme = "README.md"
exclude = [
    ".github/",
    "docs/",
    "*.md",
    ".gitignore",
    ".dockerignore",
    "Dockerfile",
    "deny.toml",
]

[[bin]]
name = "furnace"
path = "src/main.rs"


[features]
default = ["burn-import"]
wgpu = []
metal = []
cuda = []
burn-import = ["dep:burn-import"]

[dependencies]
# ML Framework
burn = { version = "0.17", features = ["train", "std", "ndarray"] }
burn-import = { version = "0.17", features = ["onnx"], optional = true }
ndarray = "0.16"

# Web Framework
axum = "0.8"
tower = { version = "0.5", features = ["limit"] }
tower-http = { version = "0.6", features = ["cors", "limit", "set-header"] }
tokio = { version = "1.46", features = ["full"] }

# CLI
clap = { version = "4.5", features = ["derive"] }

# Serialization
serde = { version = "1.0", features = ["derive"] }
serde_json = "1.0"

# Logging
tracing = "0.1"
tracing-subscriber = { version = "0.3", features = ["env-filter", "json", "chrono"] }
uuid = { version = "1.0", features = ["v4"] }

# Error Handling
thiserror = "2.0"

# Utilities
chrono = { version = "0.4", features = ["serde"] }

# Benchmarking dependencies
bollard = "0.17"  # Docker API client
plotters = "0.3"  # Chart generation
hdrhistogram = "7.5"  # High-performance histogram for latency metrics
statistical = "1.0"  # Statistical analysis
futures-util = "0.3"  # Async utilities
futures = "0.3"  # Futures executor
toml = "0.8"  # TOML configuration parsing
tempfile = "3.12"  # Temporary file handling
rand = "0.8"  # Random number generation
reqwest = { version = "0.12", features = ["json"] }  # HTTP client

[build-dependencies]
burn-import = { version = "0.17", features = ["onnx"], optional = true }

[dev-dependencies]
# Testing utilities
tempfile = "3.12"
reqwest = { version = "0.12", features = ["json"] }
tower = { version = "0.5", features = ["util"] }

# Benchmarking
criterion = { version = "0.5", features = ["html_reports"] }

# Examples dependencies
rand = "0.8"
ureq = "2.10"
protobuf = "3.4"


[[bench]]
name = "inference_benchmark"
harness = false

[profile.release]
# Maximum performance optimizations
opt-level = 3
lto = "fat"
codegen-units = 1
panic = "abort"
strip = true
debug = false
overflow-checks = false

[profile.release.build-override]
opt-level = 3
codegen-units = 1

[profile.dev]
# Faster compilation for development
opt-level = 0
debug = true
split-debuginfo = "unpacked"

[profile.test]
# Better debugging for tests
opt-level = 0
debug = true

[profile.bench]
# Optimize for benchmarks
opt-level = 3
debug = false
lto = true

# The profile that 'cargo dist' will build with
[profile.dist]
inherits = "release"
lto = "thin"