Merge branch 'main' into main

Author: 4j17h

Cargo.lock

@@ -19,7 +19,7 @@ hf-hub = { version = "0.3.2" }
 criterion = "0.5.1"
 pprof = "0.13.0"
 bindgen = "0.69.5"
-cc = "1.2.29"
+cc = "1.2.30"
 anyhow = "1.0.98"
 clap = "4.5.41"
 encoding_rs = "0.8.35"

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "embeddings"
-version = "0.1.112"
+version = "0.1.113"
 edition = "2021"
 
 [dependencies]

examples/embeddings/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "simple"
-version = "0.1.112"
+version = "0.1.113"
 edition = "2021"
 
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html

examples/simple/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "llama-cpp-2"
 description = "llama.cpp bindings for Rust"
-version = "0.1.112"
+version = "0.1.113"
 edition = "2021"
 license = "MIT OR Apache-2.0"
 repository = "https://github.com/utilityai/llama-cpp-rs"

llama-cpp-2/Cargo.toml

@@ -7,7 +7,7 @@
 //!
 //! # Examples
 //!
-//! - [simple](https://github.com/utilityai/llama-cpp-rs/tree/main/simple)
+//! - [simple](https://github.com/utilityai/llama-cpp-rs/tree/main/examples/simple)
 //!
 //! # Feature Flags
 //!

llama-cpp-2/src/lib.rs

@@ -11,7 +11,7 @@ impl LlamaTimings {
     /// Create a new `LlamaTimings`.
     /// ```
     /// # use llama_cpp_2::timing::LlamaTimings;
-    /// let timings = LlamaTimings::new(1.0, 2.0, 3.0, 4.0, 5, 6);
+    /// let timings = LlamaTimings::new(1.0, 2.0, 3.0, 4.0, 5, 6, 1);
     /// let timings_str = "load time = 2.00 ms
     /// prompt eval time = 3.00 ms / 5 tokens (0.60 ms per token, 1666.67 tokens per second)
     /// eval time = 4.00 ms / 6 runs (0.67 ms per token, 1500.00 tokens per second)\n";
@@ -26,6 +26,7 @@ impl LlamaTimings {
         t_eval_ms: f64,
         n_p_eval: i32,
         n_eval: i32,
+        n_reused: i32,
     ) -> Self {
         Self {
             timings: llama_cpp_sys_2::llama_perf_context_data {

llama-cpp-2/src/timing.rs

@@ -1,7 +1,7 @@
 [package]
 name = "llama-cpp-sys-2"
 description = "Low Level Bindings to llama.cpp"
-version = "0.1.112"
+version = "0.1.113"
 edition = "2021"
 license = "MIT OR Apache-2.0"
 repository = "https://github.com/utilityai/llama-cpp-rs"

llama-cpp-sys-2/Cargo.toml

@@ -441,6 +441,13 @@ fn main() {
     config.define("LLAMA_BUILD_TOOLS", "OFF");
     config.define("LLAMA_CURL", "OFF");
 
+    // Pass CMAKE_ environment variables down to CMake
+    for (key, value) in env::vars() {
+        if key.starts_with("CMAKE_") {
+            config.define(&key, &value);
+        }
+    }
+
     config.define(
         "BUILD_SHARED_LIBS",
         if build_shared_libs { "ON" } else { "OFF" },