[package]
edition = "2021"
name = "ferrum-cli"
version = "0.8.2"
authors = ["Ferrum Team"]
build = false
autolib = false
autobins = false
autoexamples = false
autotests = false
autobenches = false
description = "CLI for Ferrum — a Rust-native LLM inference engine"
readme = "README.md"
license = "MIT"
[features]
accelerate = ["ferrum-engine/accelerate"]
benchmark = []
candle-cuda-compat = [
"cuda",
"ferrum-engine/candle-cuda-compat",
"ferrum-kernels/candle-cuda-compat",
]
cuda = [
"ferrum-engine/cuda",
"ferrum-kernels/cuda",
"ferrum-quantization/cuda",
]
default = []
dev-tools = []
fa2-source = [
"cuda",
"ferrum-kernels/fa2-source",
]
interactive = []
marlin = ["cuda"]
metal = [
"ferrum-engine/metal",
"ferrum-kernels/metal",
"ferrum-quantization/metal",
]
native-op-artifact = [
"cuda",
"ferrum-kernels/native-op-artifact",
]
tensor-parallel = ["cuda"]
triton-kernels = [
"cuda",
"ferrum-engine/triton-kernels",
]
vllm-marlin = [
"cuda",
"ferrum-kernels/vllm-marlin",
]
vllm-moe-marlin = [
"cuda",
"ferrum-kernels/vllm-moe-marlin",
]
vllm-paged-attn-v2 = [
"cuda",
"ferrum-kernels/vllm-paged-attn-v2",
]
[lib]
name = "ferrum_cli"
path = "src/lib.rs"
[[bin]]
name = "ferrum"
path = "src/main.rs"
[[test]]
name = "chat_pty"
path = "tests/chat_pty.rs"
[[test]]
name = "chat_smoke"
path = "tests/chat_smoke.rs"
[[test]]
name = "chat_stress"
path = "tests/chat_stress.rs"
[[test]]
name = "cli_e2e"
path = "tests/cli_e2e.rs"
[[test]]
name = "quant_kl"
path = "tests/quant_kl.rs"
[[test]]
name = "reference_match"
path = "tests/reference_match.rs"
[[test]]
name = "server_agent_tools"
path = "tests/server_agent_tools.rs"
[[test]]
name = "server_lora_startup"
path = "tests/server_lora_startup.rs"
[[test]]
name = "server_openai_compat"
path = "tests/server_openai_compat.rs"
[[test]]
name = "server_prefix_cache_product"
path = "tests/server_prefix_cache_product.rs"
[[test]]
name = "server_session_cache"
path = "tests/server_session_cache.rs"
[[test]]
name = "server_smoke"
path = "tests/server_smoke.rs"
[[test]]
name = "server_stress"
path = "tests/server_stress.rs"
[[test]]
name = "server_structured_output"
path = "tests/server_structured_output.rs"
[[test]]
name = "vllm_migration_compat"
path = "tests/vllm_migration_compat.rs"
[dependencies.anyhow]
version = "1.0"
[dependencies.async-trait]
version = "0.1"
[dependencies.candle-core]
version = "0.9.2"
[dependencies.chrono]
version = "0.4"
features = ["serde"]
[dependencies.clap]
version = "4.4"
features = [
"derive",
"color",
"env",
]
[dependencies.colored]
version = "2.1"
[dependencies.config]
version = "0.14"
[dependencies.console]
version = "0.15"
[dependencies.criterion]
version = "0.5"
features = ["html_reports"]
[dependencies.dialoguer]
version = "0.11"
[dependencies.dirs]
version = "5.0"
[dependencies.ferrum-bench-core]
version = "0.8.2"
[dependencies.ferrum-engine]
version = "0.8.2"
[dependencies.ferrum-interfaces]
version = "0.8.2"
[dependencies.ferrum-kernels]
version = "0.8.2"
[dependencies.ferrum-models]
version = "0.8.2"
[dependencies.ferrum-quantization]
version = "0.8.2"
[dependencies.ferrum-scheduler]
version = "0.8.2"
[dependencies.ferrum-server]
version = "0.8.2"
[dependencies.ferrum-types]
version = "0.8.2"
[dependencies.futures]
version = "0.3"
[dependencies.indicatif]
version = "0.17"
[dependencies.libc]
version = "0.2"
[dependencies.num_cpus]
version = "1.16"
[dependencies.rand]
version = "0.9.0"
[dependencies.reqwest]
version = "0.12"
features = [
"json",
"stream",
"rustls-tls",
]
default-features = false
[dependencies.serde]
version = "1.0"
features = ["derive"]
[dependencies.serde_json]
version = "1.0"
[dependencies.serde_yaml]
version = "0.9"
[dependencies.sha2]
version = "0.10"
[dependencies.shellexpand]
version = "3.1"
[dependencies.thiserror]
version = "1.0"
[dependencies.tokenizers]
version = "0.21"
features = ["onig"]
[dependencies.tokio]
version = "1.35"
features = [
"full",
"full",
]
[dependencies.tokio-stream]
version = "0.1"
[dependencies.toml]
version = "0.8"
[dependencies.tracing]
version = "0.1"
[dependencies.tracing-subscriber]
version = "0.3"
features = [
"env-filter",
"json",
]
[dependencies.uuid]
version = "1.6"
features = [
"v4",
"serde",
]
[dependencies.walkdir]
version = "2.4"
[dev-dependencies.async-openai]
version = "0.27"
features = ["rustls"]
default-features = false
[dev-dependencies.rexpect]
version = "0.6"
[dev-dependencies.rstest]
version = "0.23"