@novastera-oss/llamarn 0.3.0 → 0.3.1

package/cpp/llama.cpp/src/unicode.cpp

@@ -557,6 +557,178 @@ static std::vector<size_t> unicode_regex_split_stl(const std::string & text, con
     return bpe_offsets;
 }
 
+// K2 system regex patterns (from tokenization_kimi.py):
+// [\p{Han}]+|[^\r\n\p{L}\p{N}]?[\p{Lu}\p{Lt}\p{Lm}\p{Lo}\p{M}&&[^\p{Han}]]*[\p{Ll}\p{Lm}\p{Lo}\p{M}&&[^\p{Han}]]+(?i:'s|'t|'re|'ve|'m|'ll|'d)?|[^\r\n\p{L}\p{N}]?[\p{Lu}\p{Lt}\p{Lm}\p{Lo}\p{M}&&[^\p{Han}]]+[\p{Ll}\p{Lm}\p{Lo}\p{M}&&[^\p{Han}]]*(?i:'s|'t|'re|'ve|'m|'ll|'d)?|\p{N}{1,3}| ?[^\s\p{L}\p{N}]+[\r\n]*|\s*[\r\n]+|\s+(?!\S)|\s+
+static std::vector<size_t> unicode_regex_split_custom_kimi_k2(const std::string & text, const std::vector<size_t> & offsets) {
+    std::vector<size_t> bpe_offsets;
+    bpe_offsets.reserve(offsets.size());
+
+    const auto cpts = unicode_cpts_from_utf8(text);
+
+    size_t start = 0;
+    for (auto offset : offsets) {
+        const size_t offset_ini = start;
+        const size_t offset_end = start + offset;
+        assert(offset_end <= cpts.size());
+        start = offset_end;
+
+        static const uint32_t OUT_OF_RANGE = 0xFFFFFFFF;
+        auto _get_cpt = [&] (const size_t pos) -> uint32_t {
+            return (offset_ini <= pos && pos < offset_end) ? cpts[pos] : OUT_OF_RANGE;
+        };
+
+        auto _get_flags = [&] (const size_t pos) -> unicode_cpt_flags {
+            return (offset_ini <= pos && pos < offset_end) ? unicode_cpt_flags_from_cpt(cpts[pos]) : unicode_cpt_flags{};
+        };
+
+        size_t _prev_end = offset_ini;
+        auto _add_token = [&] (const size_t end) -> size_t {
+            assert(_prev_end <= end && end <= offset_end);
+            size_t len = end - _prev_end;
+            if (len > 0) {
+                bpe_offsets.push_back(len);
+            }
+            _prev_end = end;
+            return len;
+        };
+
+        for (size_t pos = offset_ini; pos < offset_end; /*pos++*/ ) {
+            const uint32_t cpt = _get_cpt(pos);
+            const auto flags = _get_flags(pos);
+
+            // Pattern 1: [\p{Han}]+ (Chinese characters)
+            if (unicode_cpt_is_han(cpt)) {
+                while (unicode_cpt_is_han(_get_cpt(pos))) {
+                    pos++;
+                }
+                _add_token(pos);
+                continue;
+            }
+
+            // Pattern 2 & 3: Letter words excluding Han characters with optional contractions
+            // [^\r\n\p{L}\p{N}]?[\p{Lu}\p{Lt}\p{Lm}\p{Lo}\p{M}&&[^\p{Han}]]*[\p{Ll}\p{Lm}\p{Lo}\p{M}&&[^\p{Han}]]+(?:'s|'t|'re|'ve|'m|'ll|'d)?
+            // [^\r\n\p{L}\p{N}]?[\p{Lu}\p{Lt}\p{Lm}\p{Lo}\p{M}&&[^\p{Han}]]+[\p{Ll}\p{Lm}\p{Lo}\p{M}&&[^\p{Han}]]*(?:'s|'t|'re|'ve|'m|'ll|'d)?
+            // Check if current char is a letter OR if current char could be a leading char and next char is a letter
+            bool is_letter_pattern = (flags.is_letter && !unicode_cpt_is_han(cpt)) ||
+                                     (!(cpt == '\r' || cpt == '\n' || flags.is_letter || flags.is_number) &&
+                                      _get_flags(pos + 1).is_letter && !unicode_cpt_is_han(_get_cpt(pos + 1)));
+
+            if (is_letter_pattern) {
+                // Handle optional leading non-letter/non-number character
+                bool has_leading_char = false;
+                if (!(cpt == '\r' || cpt == '\n' || flags.is_letter || flags.is_number)) {
+                    has_leading_char = true;
+                    pos++;
+                }
+
+                // Match letter sequence (excluding Han characters)
+                bool has_letters = false;
+                while (_get_flags(pos).is_letter && !unicode_cpt_is_han(_get_cpt(pos))) {
+                    has_letters = true;
+                    pos++;
+                }
+
+                // Only proceed if we found letters (after potentially skipping leading char)
+                if (has_letters || (!has_leading_char && _get_flags(pos).is_letter && !unicode_cpt_is_han(_get_cpt(pos)))) {
+                    if (!has_letters) pos++; // consume the first letter if we didn't already
+
+                    // Continue consuming letters
+                    while (_get_flags(pos).is_letter && !unicode_cpt_is_han(_get_cpt(pos))) {
+                        pos++;
+                    }
+
+                    // Check for optional contractions (?:'s|'t|'re|'ve|'m|'ll|'d)
+                    if (_get_cpt(pos) == '\'' && pos + 1 < offset_end) {
+                        uint32_t cpt_next = unicode_tolower(_get_cpt(pos + 1));
+                        if (cpt_next == 's' || cpt_next == 't' || cpt_next == 'm' || cpt_next == 'd') {
+                            pos += 2;
+                        } else if (pos + 2 < offset_end) {
+                            uint32_t cpt_next_next = unicode_tolower(_get_cpt(pos + 2));
+                            if ((cpt_next == 'r' && cpt_next_next == 'e') ||
+                                (cpt_next == 'v' && cpt_next_next == 'e') ||
+                                (cpt_next == 'l' && cpt_next_next == 'l')) {
+                                pos += 3;
+                            }
+                        }
+                    }
+
+                    _add_token(pos);
+                    continue;
+                } else if (has_leading_char) {
+                    // We consumed a leading char but found no letters, backtrack
+                    pos--;
+                }
+            }
+
+            // Pattern 4: \p{N}{1,3} (numbers 1-3 digits)
+            if (flags.is_number) {
+                size_t ini = pos;
+                while (_get_flags(pos).is_number) {
+                    if (++pos - ini >= 3) {
+                        _add_token(pos);
+                        ini = pos;
+                    }
+                }
+                _add_token(pos);
+                continue;
+            }
+
+            // Pattern 5:  ?[^\s\p{L}\p{N}]+[\r\n]* (optional space + non-word chars + optional newlines)
+            auto flags2 = (cpt == ' ' ? _get_flags(pos + 1) : flags);
+            if (!(flags2.is_whitespace || flags2.is_letter || flags2.is_number) && flags2.as_uint()) {
+                pos += (cpt == ' ');
+                while (!(flags2.is_whitespace || flags2.is_letter || flags2.is_number) && flags2.as_uint()) {
+                    flags2 = _get_flags(++pos);
+                }
+                // Match optional [\r\n]*
+                uint32_t cpt2 = _get_cpt(pos);
+                while (cpt2 == '\r' || cpt2 == '\n') {
+                    cpt2 = _get_cpt(++pos);
+                }
+                _add_token(pos);
+                continue;
+            }
+
+            // Count whitespace characters
+            size_t num_whitespaces = 0;
+            size_t last_end_r_or_n = 0;
+            while (_get_flags(pos + num_whitespaces).is_whitespace) {
+                uint32_t cpt2 = _get_cpt(pos + num_whitespaces);
+                if (cpt2 == '\r' || cpt2 == '\n') {
+                    last_end_r_or_n = pos + num_whitespaces + 1;
+                }
+                num_whitespaces++;
+            }
+
+            // Pattern 6: \s*[\r\n]+ (whitespace with newlines)
+            if (last_end_r_or_n > 0) {
+                pos = last_end_r_or_n;
+                _add_token(pos);
+                continue;
+            }
+
+            // Pattern 7: \s+(?!\S) (trailing whitespace)
+            if (num_whitespaces > 1 && _get_cpt(pos + num_whitespaces) != OUT_OF_RANGE) {
+                pos += num_whitespaces - 1;
+                _add_token(pos);
+                continue;
+            }
+
+            // Pattern 8: \s+ (general whitespace)
+            if (num_whitespaces > 0) {
+                pos += num_whitespaces;
+                _add_token(pos);
+                continue;
+            }
+
+            // No matches - consume single character
+            _add_token(++pos);
+        }
+    }
+
+    return bpe_offsets;
+}
+
 static std::vector<size_t> unicode_regex_split_custom(const std::string & text, const std::string & regex_expr, const std::vector<size_t> & offsets) {
     std::vector<size_t> bpe_offsets;
 
@@ -567,6 +739,9 @@ static std::vector<size_t> unicode_regex_split_custom(const std::string & text,
             regex_expr == "(?:'[sS]|'[tT]|'[rR][eE]|'[vV][eE]|'[mM]|'[lL][lL]|'[dD])|[^\\r\\n\\p{L}\\p{N}]?\\p{L}+|\\p{N}{1,3}| ?[^\\s\\p{L}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+") {
 
         bpe_offsets = unicode_regex_split_custom_llama3(text, offsets);
+    } else if (regex_expr == "\\p{Han}+") {
+        // K2's first pattern - handle all K2 patterns together
+        bpe_offsets = unicode_regex_split_custom_kimi_k2(text, offsets);
     }
 
     return bpe_offsets;
@@ -672,6 +847,38 @@ uint32_t unicode_tolower(uint32_t cpt) {
     return cpt;  // Return the original code point if no lowercase mapping is found
 }
 
+bool unicode_cpt_is_han(uint32_t cpt) {
+    // Han character ranges (Chinese/CJK characters)
+    // CJK Unified Ideographs (most common)
+    if (cpt >= 0x4E00 && cpt <= 0x9FFF) return true;
+
+    // CJK Extension A
+    if (cpt >= 0x3400 && cpt <= 0x4DBF) return true;
+
+    // CJK Extension B
+    if (cpt >= 0x20000 && cpt <= 0x2A6DF) return true;
+
+    // CJK Extension C
+    if (cpt >= 0x2A700 && cpt <= 0x2B73F) return true;
+
+    // CJK Extension D
+    if (cpt >= 0x2B740 && cpt <= 0x2B81F) return true;
+
+    // CJK Extension E
+    if (cpt >= 0x2B820 && cpt <= 0x2CEAF) return true;
+
+    // CJK Extension F
+    if (cpt >= 0x2CEB0 && cpt <= 0x2EBEF) return true;
+
+    // CJK Compatibility Ideographs
+    if (cpt >= 0xF900 && cpt <= 0xFAFF) return true;
+
+    // CJK Compatibility Ideographs Supplement
+    if (cpt >= 0x2F800 && cpt <= 0x2FA1F) return true;
+
+    return false;
+}
+
 std::vector<std::string> unicode_regex_split(const std::string & text, const std::vector<std::string> & regex_exprs) {
     // unicode categories
     static const std::map<std::string, int> k_ucat_enum = {

package/cpp/llama.cpp/src/unicode.h

@@ -63,4 +63,6 @@ uint8_t     unicode_utf8_to_byte(const std::string & utf8);
 
 uint32_t unicode_tolower(uint32_t cpt);
 
+bool unicode_cpt_is_han(uint32_t cpt);
+
 std::vector<std::string> unicode_regex_split(const std::string & text, const std::vector<std::string> & regex_exprs);

package/ios/include/common.h

@@ -81,6 +81,7 @@ enum llama_example {
     LLAMA_EXAMPLE_LOOKUP,
     LLAMA_EXAMPLE_PARALLEL,
     LLAMA_EXAMPLE_TTS,
+    LLAMA_EXAMPLE_DIFFUSION,
 
     LLAMA_EXAMPLE_COUNT,
 };
@@ -177,7 +178,8 @@ struct common_params_sampling {
     std::vector<common_grammar_trigger> grammar_triggers; // optional triggers (for lazy grammars)
     std::set<llama_token>               preserved_tokens;
 
-    std::vector<llama_logit_bias> logit_bias; // logit biases to apply
+    std::vector<llama_logit_bias> logit_bias;     // logit biases to apply
+    std::vector<llama_logit_bias> logit_bias_eog; // pre-calculated logit biases for EOG tokens
 
     // print the parameters into a string
     std::string print() const;
@@ -217,6 +219,14 @@ struct common_params_vocoder {
     bool use_guide_tokens = false; // enable guide tokens to improve TTS accuracy            // NOLINT
 };
 
+struct common_params_diffusion {
+    int32_t steps       = 64;     // number of diffusion steps
+    float   eps         = 1e-3f;  // epsilon for timesteps
+    int32_t algorithm   = 0;      // diffusion algorithm (0=ORIGIN, 1=MASKGIT_PLUS, 2=TOPK_MARGIN, 3=ENTROPY)
+    float   alg_temp    = 0.0f;   // algorithm temperature
+    bool    visual_mode = false;  // show progressive diffusion on screen
+};
+
 enum common_reasoning_format {
     COMMON_REASONING_FORMAT_NONE,
     COMMON_REASONING_FORMAT_DEEPSEEK_LEGACY, // Extract thinking tag contents and return as `message.reasoning_content`, or leave inline in <think> tags in stream mode
@@ -268,6 +278,7 @@ struct common_params {
     struct common_params_sampling    sampling;
     struct common_params_speculative speculative;
     struct common_params_vocoder     vocoder;
+    struct common_params_diffusion   diffusion;
 
     struct common_params_model model;
 
@@ -330,6 +341,7 @@ struct common_params {
     bool no_perf           = false; // disable performance metrics
     bool ctx_shift         = true;  // context shift on inifinite text generation
     bool swa_full          = false; // use full-size SWA cache (https://github.com/ggml-org/llama.cpp/pull/13194#issuecomment-2868343055)
+    bool kv_unified        = false; // enable unified KV cache
 
     bool input_prefix_bos  = false; // prefix BOS to user inputs, preceding input_prefix
     bool use_mmap          = true;  // use mmap for faster loads
@@ -420,9 +432,10 @@ struct common_params {
     int32_t n_save_freq =  0; // save the imatrix every n_save_freq iterations
     int32_t i_chunk     =  0; // start processing from this chunk
 
-    bool process_output = false; // collect data for the output tensor
-    bool compute_ppl    = true;  // whether to compute perplexity
-    bool parse_special  = false; // whether to parse special tokens during imatrix tokenization
+    bool process_output  = false; // collect data for the output tensor
+    bool compute_ppl     = true;  // whether to compute perplexity
+    bool show_statistics = false; // show imatrix statistics per tensor
+    bool parse_special   = false; // whether to parse special tokens during imatrix tokenization
 
     // cvector-generator params
     int n_pca_batch = 100;
@@ -522,6 +535,7 @@ static bool string_starts_with(const std::string & str,
 
 // While we wait for C++20's std::string::ends_with...
 bool string_ends_with(const std::string_view & str, const std::string_view & suffix);
+bool string_remove_suffix(std::string & str, const std::string_view & suffix);
 size_t string_find_partial_stop(const std::string_view & str, const std::string_view & stop);
 
 bool string_parse_kv_override(const char * data, std::vector<llama_model_kv_override> & overrides);

package/ios/include/llama.h

@@ -71,12 +71,13 @@ extern "C" {
     typedef int32_t llama_seq_id;
 
     enum llama_vocab_type {
-        LLAMA_VOCAB_TYPE_NONE = 0, // For models without vocab
-        LLAMA_VOCAB_TYPE_SPM  = 1, // LLaMA tokenizer based on byte-level BPE with byte fallback
-        LLAMA_VOCAB_TYPE_BPE  = 2, // GPT-2 tokenizer based on byte-level BPE
-        LLAMA_VOCAB_TYPE_WPM  = 3, // BERT tokenizer based on WordPiece
-        LLAMA_VOCAB_TYPE_UGM  = 4, // T5 tokenizer based on Unigram
-        LLAMA_VOCAB_TYPE_RWKV = 5, // RWKV tokenizer based on greedy tokenization
+        LLAMA_VOCAB_TYPE_NONE   = 0, // For models without vocab
+        LLAMA_VOCAB_TYPE_SPM    = 1, // LLaMA tokenizer based on byte-level BPE with byte fallback
+        LLAMA_VOCAB_TYPE_BPE    = 2, // GPT-2 tokenizer based on byte-level BPE
+        LLAMA_VOCAB_TYPE_WPM    = 3, // BERT tokenizer based on WordPiece
+        LLAMA_VOCAB_TYPE_UGM    = 4, // T5 tokenizer based on Unigram
+        LLAMA_VOCAB_TYPE_RWKV   = 5, // RWKV tokenizer based on greedy tokenization
+        LLAMA_VOCAB_TYPE_PLAMO2 = 6, // PLaMo-2 tokenizer based on Aho-Corasick with dynamic programming
     };
 
     enum llama_rope_type {
@@ -334,6 +335,9 @@ extern "C" {
         bool swa_full;    // use full-size SWA cache (https://github.com/ggml-org/llama.cpp/pull/13194#issuecomment-2868343055)
                           // NOTE: setting to false when n_seq_max > 1 can cause bad performance in some cases
                           //       ref: https://github.com/ggml-org/llama.cpp/pull/13845#issuecomment-2924800573
+        bool kv_unified;  // use a unified buffer across the input sequences when computing the attention
+                          // try to disable when n_seq_max > 1 for improved performance when the sequences do not share a large prefix
+                          // ref: https://github.com/ggml-org/llama.cpp/pull/14363
     };
 
     // model quantization parameters
@@ -724,7 +728,7 @@ extern "C" {
     //   - lazily on next llama_decode()
     // p0 < 0 : [0,  p1]
     // p1 < 0 : [p0, inf)
-    DEPRECATED(void llama_kv_self_seq_div(
+    DEPRECATED(LLAMA_API void llama_kv_self_seq_div(
             struct llama_context * ctx,
                     llama_seq_id   seq_id,
                        llama_pos   p0,
@@ -952,6 +956,7 @@ extern "C" {
     // in the order they have appeared in the batch.
     // Rows: number of tokens for which llama_batch.logits[i] != 0
     // Cols: n_vocab
+    // TODO: deprecate in favor of llama_get_logits_ith() (ref: https://github.com/ggml-org/llama.cpp/pull/14853#issuecomment-3113143522)
     LLAMA_API float * llama_get_logits(struct llama_context * ctx);
 
     // Logits for the ith token. For positive indices, Equivalent to:
@@ -966,6 +971,7 @@ extern "C" {
     // in the order they have appeared in the batch.
     // shape: [n_outputs*n_embd]
     // Otherwise, returns NULL.
+    // TODO: deprecate in favor of llama_get_embeddings_ith() (ref: https://github.com/ggml-org/llama.cpp/pull/14853#issuecomment-3113143522)
     LLAMA_API float * llama_get_embeddings(struct llama_context * ctx);
 
     // Get the embeddings for the ith token. For positive indices, Equivalent to:
@@ -1004,6 +1010,7 @@ extern "C" {
     LLAMA_API llama_token llama_vocab_sep(const struct llama_vocab * vocab); // sentence separator
     LLAMA_API llama_token llama_vocab_nl (const struct llama_vocab * vocab); // next-line
     LLAMA_API llama_token llama_vocab_pad(const struct llama_vocab * vocab); // padding
+    LLAMA_API llama_token llama_vocab_mask(const struct llama_vocab * vocab); // mask
 
     LLAMA_API bool llama_vocab_get_add_bos(const struct llama_vocab * vocab);
     LLAMA_API bool llama_vocab_get_add_eos(const struct llama_vocab * vocab);
@@ -1389,6 +1396,7 @@ extern "C" {
 
         int32_t n_p_eval;
         int32_t n_eval;
+        int32_t n_reused; // number of times a ggml compute graph had been reused
     };
 
     struct llama_perf_sampler_data {

package/ios/libs/llama.xcframework/Info.plist

@@ -10,7 +10,7 @@
 			<key>DebugSymbolsPath</key>
 			<string>dSYMs</string>
 			<key>LibraryIdentifier</key>
-			<string>xros-arm64_x86_64-simulator</string>
+			<string>ios-arm64_x86_64-simulator</string>
 			<key>LibraryPath</key>
 			<string>llama.framework</string>
 			<key>SupportedArchitectures</key>
@@ -19,7 +19,7 @@
 				<string>x86_64</string>
 			</array>
 			<key>SupportedPlatform</key>
-			<string>xros</string>
+			<string>ios</string>
 			<key>SupportedPlatformVariant</key>
 			<string>simulator</string>
 		</dict>
@@ -45,7 +45,7 @@
 			<key>DebugSymbolsPath</key>
 			<string>dSYMs</string>
 			<key>LibraryIdentifier</key>
-			<string>tvos-arm64</string>
+			<string>xros-arm64</string>
 			<key>LibraryPath</key>
 			<string>llama.framework</string>
 			<key>SupportedArchitectures</key>
@@ -53,7 +53,7 @@
 				<string>arm64</string>
 			</array>
 			<key>SupportedPlatform</key>
-			<string>tvos</string>
+			<string>xros</string>
 		</dict>
 		<dict>
 			<key>BinaryPath</key>
@@ -61,26 +61,23 @@
 			<key>DebugSymbolsPath</key>
 			<string>dSYMs</string>
 			<key>LibraryIdentifier</key>
-			<string>ios-arm64_x86_64-simulator</string>
+			<string>tvos-arm64</string>
 			<key>LibraryPath</key>
 			<string>llama.framework</string>
 			<key>SupportedArchitectures</key>
 			<array>
 				<string>arm64</string>
-				<string>x86_64</string>
 			</array>
 			<key>SupportedPlatform</key>
-			<string>ios</string>
-			<key>SupportedPlatformVariant</key>
-			<string>simulator</string>
+			<string>tvos</string>
 		</dict>
 		<dict>
 			<key>BinaryPath</key>
-			<string>llama.framework/Versions/A/llama</string>
+			<string>llama.framework/llama</string>
 			<key>DebugSymbolsPath</key>
 			<string>dSYMs</string>
 			<key>LibraryIdentifier</key>
-			<string>macos-arm64_x86_64</string>
+			<string>xros-arm64_x86_64-simulator</string>
 			<key>LibraryPath</key>
 			<string>llama.framework</string>
 			<key>SupportedArchitectures</key>
@@ -89,23 +86,26 @@
 				<string>x86_64</string>
 			</array>
 			<key>SupportedPlatform</key>
-			<string>macos</string>
+			<string>xros</string>
+			<key>SupportedPlatformVariant</key>
+			<string>simulator</string>
 		</dict>
 		<dict>
 			<key>BinaryPath</key>
-			<string>llama.framework/llama</string>
+			<string>llama.framework/Versions/A/llama</string>
 			<key>DebugSymbolsPath</key>
 			<string>dSYMs</string>
 			<key>LibraryIdentifier</key>
-			<string>xros-arm64</string>
+			<string>macos-arm64_x86_64</string>
 			<key>LibraryPath</key>
 			<string>llama.framework</string>
 			<key>SupportedArchitectures</key>
 			<array>
 				<string>arm64</string>
+				<string>x86_64</string>
 			</array>
 			<key>SupportedPlatform</key>
-			<string>xros</string>
+			<string>macos</string>
 		</dict>
 		<dict>
 			<key>BinaryPath</key>