llama_cpp 0.4.0 → 0.5.1

data/ext/llama_cpp/src/llama.h

@@ -10,6 +10,7 @@
 #endif // GGML_USE_CUBLAS
 #include <stddef.h>
 #include <stdint.h>
+#include <stdio.h>
 #include <stdbool.h>
 
 #ifdef LLAMA_SHARED
@@ -163,6 +164,7 @@ extern "C" {
         enum llama_ftype ftype;      // quantize to this llama_ftype
         bool allow_requantize;       // allow quantizing non-f32/f16 tensors
         bool quantize_output_tensor; // quantize output.weight
+        bool only_copy;              // only copy tensors - ftype, allow_requantize and quantize_output_tensor are ignored
     } llama_model_quantize_params;
 
     // grammar types
@@ -254,7 +256,11 @@ extern "C" {
     LLAMA_API int llama_model_n_embd (const struct llama_model * model);
 
     // Get a string describing the model type
-    LLAMA_API int llama_model_type(const struct llama_model * model, char * buf, size_t buf_size);
+    LLAMA_API int llama_model_desc(const struct llama_model * model, char * buf, size_t buf_size);
+    // Returns the total size of all the tensors in the model in bytes
+    LLAMA_API uint64_t llama_model_size(const struct llama_model * model);
+    // Returns the total number of parameters in the model
+    LLAMA_API uint64_t llama_model_n_params(const struct llama_model * model);
 
     // Returns 0 on success
     LLAMA_API int llama_model_quantize(
@@ -377,15 +383,17 @@ extern "C" {
                              int   n_max_tokens,
                             bool   add_bos);
 
-    // Token Id -> String. Uses the vocabulary in the provided context
-    // Does not write null terminator to the buffer
-    LLAMA_API int llama_token_to_str(
+    // Token Id -> Piece.
+    // Uses the vocabulary in the provided context.
+    // Does not write null terminator to the buffer.
+    // User code is responsible to remove the leading whitespace of the first non-BOS token when decoding multiple tokens.
+    LLAMA_API int llama_token_to_piece(
             const struct llama_context * ctx,
                            llama_token   token,
                                   char * buf,
                                   int    length);
 
-    LLAMA_API int llama_token_to_str_with_model(
+    LLAMA_API int llama_token_to_piece_with_model(
               const struct llama_model * model,
                            llama_token   token,
                                   char * buf,
@@ -402,6 +410,8 @@ extern "C" {
 
     LLAMA_API void llama_grammar_free(struct llama_grammar * grammar);
 
+    LLAMA_API struct llama_grammar * llama_grammar_copy(const struct llama_grammar * grammar);
+
     //
     // Sampling functions
     //
@@ -465,6 +475,43 @@ extern "C" {
     /// @details Accepts the sampled token into the grammar
     LLAMA_API void llama_grammar_accept_token(struct llama_context * ctx, struct llama_grammar * grammar, llama_token token);
 
+    //
+    // Beam search
+    //
+
+    struct llama_beam_view {
+        const llama_token * tokens;
+        size_t n_tokens;
+        float p;   // Cumulative beam probability (renormalized relative to all beams)
+        bool eob;  // Callback should set this to true when a beam is at end-of-beam.
+    };
+
+    // Passed to beam_search_callback function.
+    // Whenever 0 < common_prefix_length, this number of tokens should be copied from any of the beams
+    // (e.g. beams[0]) as they will be removed (shifted) from all beams in all subsequent callbacks.
+    // These pointers are valid only during the synchronous callback, so should not be saved.
+    struct llama_beams_state {
+        struct llama_beam_view * beam_views;
+        size_t n_beams;               // Number of elements in beam_views[].
+        size_t common_prefix_length;  // Current max length of prefix tokens shared by all beams.
+        bool last_call;               // True iff this is the last callback invocation.
+    };
+
+    // Type of pointer to the beam_search_callback function.
+    // void* callback_data is any custom data passed to llama_beam_search, that is subsequently
+    // passed back to beam_search_callback. This avoids having to use global variables in the callback.
+    typedef void (*llama_beam_search_callback_fn_t)(void * callback_data, struct llama_beams_state);
+
+    /// @details Deterministically returns entire sentence constructed by a beam search.
+    /// @param ctx Pointer to the llama_context.
+    /// @param callback Invoked for each iteration of the beam_search loop, passing in beams_state.
+    /// @param callback_data A pointer that is simply passed back to callback.
+    /// @param n_beams Number of beams to use.
+    /// @param n_past Number of tokens already evaluated.
+    /// @param n_predict Maximum number of tokens to predict. EOS may occur earlier.
+    /// @param n_threads Number of threads as passed to llama_eval().
+    LLAMA_API void llama_beam_search(struct llama_context * ctx, llama_beam_search_callback_fn_t callback, void * callback_data, size_t n_beams, int n_past, int n_predict, int n_threads);
+
     // Performance information
     LLAMA_API struct llama_timings llama_get_timings(struct llama_context * ctx);
     LLAMA_API void llama_print_timings(struct llama_context * ctx);
@@ -477,6 +524,8 @@ extern "C" {
     // If this is not called, or NULL is supplied, everything is output on stderr.
     LLAMA_API void llama_log_set(llama_log_callback log_callback, void * user_data);
 
+    LLAMA_API void llama_dump_timing_info_yaml(FILE * stream, const struct llama_context * ctx);
+
 #ifdef __cplusplus
 }
 #endif

data/lib/llama_cpp/version.rb

@@ -3,8 +3,8 @@
 # llama_cpp.rb provides Ruby bindings for the llama.cpp.
 module LLaMACpp
   # The version of llama_cpp.rb you install.
-  VERSION = '0.4.0'
+  VERSION = '0.5.1'
 
   # The version of llama.cpp bundled with llama_cpp.rb.
-  LLAMA_CPP_VERSION = 'b1060'
+  LLAMA_CPP_VERSION = 'b1198'
 end

data/lib/llama_cpp.rb

@@ -99,7 +99,7 @@ module LLaMACpp
         end
       end
 
-      embd.each { |token| output << context.token_to_str(token) }
+      embd.each { |token| output << context.token_to_piece(token) }
 
       break if !embd.empty? && embd[-1] == context.token_eos
     end

data/sig/llama_cpp.rbs

@@ -76,9 +76,11 @@ module LLaMACpp
     def n_vocab: () -> Integer
     def n_ctx: () -> Integer
     def n_embd: () -> Integer
-    def token_to_str: (Integer) -> String
+    def token_to_piece: (Integer) -> String
     def tokenize: (text: String, ?n_max_tokens: Integer, ?add_bos: bool) -> Array[Integer]
-    def type: () -> String
+    def desc: () -> String
+    def size: () -> Integer
+    def n_params: () -> Integer
   end
 
   class Timings
@@ -116,7 +118,7 @@ module LLaMACpp
     def timings: () -> ::LLaMACpp::Timings
     def print_timings: () -> void
     def reset_timings: () -> void
-    def token_to_str: (Integer) -> String
+    def token_to_piece: (Integer) -> String
     def tokenize: (text: String, ?n_max_tokens: Integer, ?add_bos: bool) -> Array[Integer]
     def kv_cache_token_count: () -> Integer
     def set_rng_seed: (Integer) -> void

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: llama_cpp
 version: !ruby/object:Gem::Version
-  version: 0.4.0
+  version: 0.5.1
 platform: ruby
 authors:
 - yoshoku
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2023-08-26 00:00:00.000000000 Z
+date: 2023-09-08 00:00:00.000000000 Z
 dependencies: []
 description: llama_cpp.rb provides Ruby bindings for the llama.cpp.
 email: