reflex-cache 0.1.2

Episodic memory and high-speed semantic cache for LLM responses
[[bin]]
name = "reflex"
path = "src/main.rs"

[dependencies.anyhow]
version = "1.0"

[dependencies.async-openai]
features = ["chat-completion-types"]
version = "0.31"

[dependencies.async-trait]
version = "0.1.89"

[dependencies.axum]
version = "0.8"

[dependencies.bitvec]
version = "1.0.1"

[dependencies.blake3]
version = "1.6"

[dependencies.bytemuck]
version = "1.19"

[dependencies.candle-core]
version = "0.9"

[dependencies.candle-nn]
version = "0.9"

[dependencies.candle-transformers]
version = "0.9"

[dependencies.chrono]
version = "0.4.42"

[dependencies.futures-util]
version = "0.3.31"

[dependencies.genai]
version = "0.4.4"

[dependencies.half]
features = ["bytemuck"]
version = "2.4"

[dependencies.hyper]
features = ["full"]
version = "1.0"

[dependencies.memmap2]
version = "0.9"

[dependencies.mimalloc]
default-features = false
version = "0.1"

[dependencies.moka]
features = ["sync"]
version = "0.12.11"

[dependencies.parking_lot]
version = "0.12"

[dependencies.qdrant-client]
version = "1.13"

[dependencies.reqwest]
features = ["json"]
version = "0.12"

[dependencies.rkyv]
features = ["bytecheck"]
version = "0.8"

[dependencies.serde]
features = ["derive"]
version = "1.0"

[dependencies.serde_json]
version = "1.0"

[dependencies.tauq]
version = "0.1.0"

[dependencies.tempfile]
version = "3.15"

[dependencies.thiserror]
version = "2.0"

[dependencies.tokenizers]
version = "0.22"

[dependencies.tokio]
features = ["full"]
version = "1.48"

[dependencies.tokio-stream]
version = "0.1.17"

[dependencies.tower]
version = "0.5"

[dependencies.tower-http]
features = ["trace", "cors"]
version = "0.6"

[dependencies.tracing]
version = "0.1"

[dependencies.tracing-subscriber]
features = ["env-filter"]
version = "0.3"

[dependencies.uuid]
features = ["v4", "fast-rng"]
version = "1.19.0"

[dev-dependencies.criterion]
version = "0.8"

[dev-dependencies.futures]
version = "0.3"

[dev-dependencies.http-body-util]
version = "0.1"

[dev-dependencies.hyper]
features = ["full"]
version = "1.0"

[dev-dependencies.serial_test]
version = "3.2"

[dev-dependencies.tempfile]
version = "3.15"

[dev-dependencies.tokio-test]
version = "0.4"

[dev-dependencies.tower]
features = ["util"]
version = "0.5"

[features]
cpu = []
cuda = ["candle-core/cuda", "candle-transformers/cuda"]
default = []
metal = ["candle-core/metal", "candle-transformers/metal"]
mock = []

[lib]
name = "reflex"
path = "src/lib.rs"

[package]
autobenches = false
autobins = false
autoexamples = false
autolib = false
autotests = false
build = false
categories = ["caching", "database"]
description = "Episodic memory and high-speed semantic cache for LLM responses"
documentation = "https://docs.rs/reflex-cache"
edition = "2024"
keywords = ["llm", "cache", "semantic", "embedding", "ai"]
license = "AGPL-3.0"
name = "reflex-cache"
readme = "README.md"
repository = "https://github.com/ccheney/reflex"
rust-version = "1.92"
version = "0.1.2"

[package.metadata.docs.rs]
all-features = true
rustdoc-args = ["--cfg", "docsrs"]

[[test]]
name = "benchmark_real"
path = "tests/benchmark_real.rs"

[[test]]
name = "coding_patterns_tests"
path = "tests/coding_patterns_tests.rs"

[[test]]
name = "e2e_tests"
path = "tests/e2e_tests.rs"

[[test]]
name = "hashing_integration"
path = "tests/hashing_integration.rs"

[[test]]
name = "integration_real"
path = "tests/integration_real.rs"

[[test]]
name = "knowledge_graph_tests"
path = "tests/knowledge_graph_tests.rs"

[[test]]
name = "qdrant_mock"
path = "tests/qdrant_mock.rs"

[[test]]
name = "request_shape_tests"
path = "tests/request_shape_tests.rs"

[[test]]
name = "reranker_tests"
path = "tests/reranker_tests.rs"

[[test]]
name = "storage_integration"
path = "tests/storage_integration.rs"