mirror of
https://github.com/whit3rabbit/anyllm-proxy.git
synced 2026-09-22 16:00:51 +00:00
whit3rabbitandClaude Opus 4.8
ff03a7ebd7
Move Unreleased changelog entries (RTK compression, opt-in prompt optimizer, release binary archives, Bedrock allowlist fix, model discovery fix) into the 0.14.0 section. Bump workspace + inter-crate versions 0.13.0 -> 0.14.0 and the README deb filename. Fix clippy missing_const_for_thread_local in optimize-core alloc_budget test (const-init the thread-locals) so cargo clippy -D warnings is clean. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
42 lines
1.7 KiB
TOML
42 lines
1.7 KiB
TOML
[package]
|
|
name = "anyllm_optimize_cli"
|
|
description = "Offline runner for FFEC: token-usage/savings eval harness, parity, shadow-diff"
|
|
version.workspace = true
|
|
edition.workspace = true
|
|
license.workspace = true
|
|
repository.workspace = true
|
|
|
|
[[bin]]
|
|
name = "optimize-eval"
|
|
path = "src/main.rs"
|
|
|
|
# Opt-in prerequisite fetcher for the `onnx` scorer. Builds without the heavy ONNX
|
|
# Runtime stack (download + sha256 verify only); running it is an explicit operator step,
|
|
# never automatic. See src/model_fetch.rs.
|
|
[[bin]]
|
|
name = "optimize-model"
|
|
path = "src/model_fetch.rs"
|
|
|
|
[features]
|
|
default = []
|
|
# M3.6: wire the real LLMLingua2Pass scorer (--llmlingua2-model-dir) into the harness
|
|
# behind the frontier, instead of always defaulting to UniformScorer. Pulls in the same
|
|
# heavy ONNX Runtime deps as `anyllm_optimize_scorer/onnx` — opt-in only.
|
|
onnx = ["anyllm_optimize_scorer/onnx"]
|
|
|
|
# M3.6: exact OpenAI token counting for the eval harness's reported est_raw/est_comp
|
|
# columns (ALGO.md: tiktoken o200k_base for OpenAI; bytes/3.6 heuristic elsewhere, since
|
|
# Anthropic's tokenizer is unpublished). Opt-in only — pulls in tiktoken-rs's BPE tables.
|
|
tiktoken = ["dep:tiktoken-rs"]
|
|
|
|
[dependencies]
|
|
anyllm_optimize_core = { path = "../crates/optimize-core", version = "0.14.0" }
|
|
anyllm_optimize_passes = { path = "../crates/optimize-passes", version = "0.14.0" }
|
|
anyllm_optimize_scorer = { path = "../crates/optimize-scorer", version = "0.14.0" }
|
|
serde = { version = "1", features = ["derive"] }
|
|
serde_json = "1"
|
|
clap = { version = "4", features = ["derive", "env"] }
|
|
anyhow = "1"
|
|
reqwest = { version = "0.12", default-features = false, features = ["blocking", "json", "rustls-tls"] }
|
|
tiktoken-rs = { version = "0.9", optional = true }
|