[package]
name = "eoc-eval"
version.workspace = true
edition.workspace = true
rust-version.workspace = true
license.workspace = true
repository.workspace = true
authors.workspace = true
description = "EOC — canonical LLM evaluation harnesses (MMLU, GPQA, HumanEval, BBH, IFEval, ...) with joules-per-correct."

[features]
default = []
download = ["dep:reqwest"]
python = ["dep:pyo3"]

[dependencies]
eoc-core.workspace = true
eoc-neural.workspace = true
eoc-bench-runner = { path = "../../../eoc-bench/crates/eoc-bench-runner" }
tokio = { workspace = true, features = ["macros", "rt", "sync", "fs", "process", "io-util", "time"] }
async-trait.workspace = true
serde = { workspace = true }
serde_json = "1"
chrono = { version = "0.4", default-features = false, features = ["clock", "serde"] }
thiserror.workspace = true
tracing = "0.1"
regex = "1"
reqwest = { version = "0.13", default-features = false, features = ["rustls", "json"], optional = true }
pyo3 = { version = "0.28", features = ["auto-initialize"], optional = true }

[dev-dependencies]
tokio = { workspace = true, features = ["macros", "rt-multi-thread", "test-util"] }