cactus-react-native 0.0.1

package/ios/cactus.xcframework/tvos-arm64_x86_64-simulator/cactus.framework/Headers/sampling.h

@@ -0,0 +1,107 @@
+#pragma once
+
+#include "llama.h"
+
+#include "common.h"
+
+#include <string>
+#include <vector>
+
+// common_sampler extends llama_sampler with additional functionality:
+//
+//  - grammar support
+//  - custom sampler logic based on the parameters
+//  - history of the last accepted tokens
+//  - performance metrics
+//
+// This goal is to have a common implementation of the sampling logic shared across the examples.
+// For example, depending on the temperature, the sampling chain can be very simple (greedy) or more
+// complex (top-k, top-p, etc).
+//
+// Another example is related to the grammar. In general, the grammar constraints applied on the full
+// vocabulary can be very taxing. To improve performance, the grammar can be applied only to the sampled
+// token in order to verify if it fits the grammar. And only if the token doesn't fit the grammar, the
+// grammar constraints are applied to the full vocabulary and the token is resampled.
+//
+// The common_sampler also maintains a container with the last accepted tokens. In the future, this can
+// be moved into the core llama library.
+//
+// For convenience, the common_sampler also maintains a container with the current candidate tokens.
+// This can be used to access the probabilities of the rest of the non-sampled tokens.
+//
+// TODO: measure grammar performance
+//
+
+struct common_sampler;
+
+// llama_sampler API overloads
+
+struct common_sampler * common_sampler_init(const struct llama_model * model, const struct common_params_sampling & params);
+
+void common_sampler_free(struct common_sampler * gsmpl);
+
+// if accept_grammar is true, the token is accepted both by the sampling chain and the grammar
+void                    common_sampler_accept(struct common_sampler * gsmpl, llama_token token, bool accept_grammar);
+void                    common_sampler_reset (struct common_sampler * gsmpl);
+struct common_sampler * common_sampler_clone (struct common_sampler * gsmpl);
+
+// arguments can be nullptr to skip printing
+void common_perf_print(const struct llama_context * ctx, const struct common_sampler * gsmpl);
+
+// extended sampling implementation:
+//
+// - set logits
+// - apply the configured sampler chain
+// - check if the token fits the grammar (if any)
+// - if not: resample by first applying the grammar constraints and then sampling again (slower path)
+//
+// if grammar_first is true, the grammar is applied before the samplers (slower)
+// useful in cases where all the resulting candidates (not just the sampled one) must fit the grammar
+//
+llama_token common_sampler_sample(struct common_sampler * gsmpl, struct llama_context * ctx, int idx, bool grammar_first = false);
+
+// generalized version of common_sampler_sample
+//
+// will cross-reference the sampled tokens with a batch of draft tokens and accept those that match
+// if the sampler disagrees at some point, we stop and return the accepted tokens up to now
+//
+//      common_sampler_sample_n(gsmpl, ctx, { idx }, {});
+//
+// is equivalent to
+//
+//      common_sampler_sample(gsmpl, ctx, idx);
+//      common_sampler_accept(gsmpl, token, true);
+//
+// requires: idxs.size() == draft.size() + 1
+//
+// returns at least 1 token, up to idxs.size()
+//
+std::vector<llama_token> common_sampler_sample_and_accept_n(struct common_sampler * gsmpl, struct llama_context * ctx, const std::vector<int> & idxs, const llama_tokens & draft, bool grammar_first = false);
+
+// assume idxs == [ 0, 1, 2, ..., draft.size() ]
+std::vector<llama_token> common_sampler_sample_and_accept_n(struct common_sampler * gsmpl, struct llama_context * ctx, const llama_tokens & draft, bool grammar_first = false);
+
+uint32_t common_sampler_get_seed(const struct common_sampler * gsmpl);
+
+// helpers
+
+// access the internal list of current candidate tokens
+llama_token_data_array * common_sampler_get_candidates(struct common_sampler * gsmpl);
+
+// get the last accepted token
+llama_token common_sampler_last(const struct common_sampler * gsmpl);
+
+// print the sampler chain into a string
+std::string common_sampler_print(const struct common_sampler * gsmpl);
+
+// get a string representation of the last accepted tokens
+std::string common_sampler_prev_str(common_sampler * gsmpl, llama_context * ctx, int n);
+
+char        common_sampler_type_to_chr(enum common_sampler_type cnstr);
+std::string common_sampler_type_to_str(enum common_sampler_type cnstr);
+
+std::vector<enum common_sampler_type> common_sampler_types_from_names(const std::vector<std::string> & names, bool allow_alt_names);
+std::vector<enum common_sampler_type> common_sampler_types_from_chars(const std::string & chars);
+
+llama_sampler * llama_sampler_init_llg(const llama_vocab * vocab,
+                const char * grammar_kind, const char * grammar_data);

package/ios/cactus.xcframework/tvos-arm64_x86_64-simulator/cactus.framework/Headers/sgemm.h

@@ -0,0 +1,14 @@
+#pragma once
+#include <stdint.h>
+#include <stdbool.h>
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+bool llamafile_sgemm(const struct lm_ggml_compute_params * params, int64_t, int64_t, int64_t,
+                     const void *, int64_t, const void *, int64_t, void *, int64_t,
+                     int, int, int);
+
+#ifdef __cplusplus
+}
+#endif

package/ios/cactus.xcframework/tvos-arm64_x86_64-simulator/cactus.framework/Headers/unicode-data.h

@@ -0,0 +1,20 @@
+#pragma once
+
+#include <cstdint>
+#include <vector>
+#include <unordered_map>
+#include <unordered_set>
+
+struct range_nfd {
+    uint32_t first;
+    uint32_t last;
+    uint32_t nfd;
+};
+
+static const uint32_t MAX_CODEPOINTS = 0x110000;
+
+extern const std::initializer_list<std::pair<uint32_t, uint16_t>> unicode_ranges_flags;
+extern const std::unordered_set<uint32_t> unicode_set_whitespace;
+extern const std::initializer_list<std::pair<uint32_t, uint32_t>> unicode_map_lowercase;
+extern const std::initializer_list<std::pair<uint32_t, uint32_t>> unicode_map_uppercase;
+extern const std::initializer_list<range_nfd> unicode_ranges_nfd;

package/ios/cactus.xcframework/tvos-arm64_x86_64-simulator/cactus.framework/Headers/unicode.h

@@ -0,0 +1,66 @@
+#pragma once
+
+#include <cstdint>
+#include <string>
+#include <vector>
+
+struct unicode_cpt_flags {
+    enum {
+        UNDEFINED       = 0x0001,
+        NUMBER          = 0x0002,  // regex: \p{N}
+        LETTER          = 0x0004,  // regex: \p{L}
+        SEPARATOR       = 0x0008,  // regex: \p{Z}
+        ACCENT_MARK     = 0x0010,  // regex: \p{M}
+        PUNCTUATION     = 0x0020,  // regex: \p{P}
+        SYMBOL          = 0x0040,  // regex: \p{S}
+        CONTROL         = 0x0080,  // regex: \p{C}
+        MASK_CATEGORIES = 0x00FF,
+    };
+
+    // codepoint type
+    uint16_t is_undefined   : 1;
+    uint16_t is_number      : 1;  // regex: \p{N}
+    uint16_t is_letter      : 1;  // regex: \p{L}
+    uint16_t is_separator   : 1;  // regex: \p{Z}
+    uint16_t is_accent_mark : 1;  // regex: \p{M}
+    uint16_t is_punctuation : 1;  // regex: \p{P}
+    uint16_t is_symbol      : 1;  // regex: \p{S}
+    uint16_t is_control     : 1;  // regex: \p{C}
+    // helper flags
+    uint16_t is_whitespace  : 1;  // regex: \s
+    uint16_t is_lowercase   : 1;
+    uint16_t is_uppercase   : 1;
+    uint16_t is_nfd         : 1;
+
+    // decode from uint16
+    inline unicode_cpt_flags(const uint16_t flags = 0) {
+        *reinterpret_cast<uint16_t*>(this) = flags;
+    }
+
+    inline uint16_t as_uint() const {
+        return *reinterpret_cast<const uint16_t*>(this);
+    }
+
+    inline uint16_t category_flag() const {
+        return this->as_uint() & MASK_CATEGORIES;
+    }
+};
+
+size_t unicode_len_utf8(char src);
+
+std::string unicode_cpt_to_utf8  (uint32_t cpt);
+uint32_t    unicode_cpt_from_utf8(const std::string & utf8, size_t & offset);
+
+std::vector<uint32_t> unicode_cpts_from_utf8(const std::string & utf8);
+
+std::vector<uint32_t> unicode_cpts_normalize_nfd(const std::vector<uint32_t> & cpts);
+
+unicode_cpt_flags unicode_cpt_flags_from_cpt (uint32_t cpt);
+unicode_cpt_flags unicode_cpt_flags_from_utf8(const std::string & utf8);
+
+std::string unicode_byte_to_utf8(uint8_t byte);
+uint8_t     unicode_utf8_to_byte(const std::string & utf8);
+
+uint32_t unicode_tolower(uint32_t cpt);
+
+std::vector<std::string> unicode_regex_split(const std::string & text, const std::vector<std::string> & regex_exprs);

package/ios/cactus.xcframework/tvos-arm64_x86_64-simulator/cactus.framework/_CodeSignature/CodeResources

@@ -0,0 +1,101 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
+<plist version="1.0">
+<dict>
+	<key>files</key>
+	<dict>
+		<key>Info.plist</key>
+		<data>
+		TxD6NsHbP5Sud08syz/nlf2bmVw=
+		</data>
+	</dict>
+	<key>files2</key>
+	<dict/>
+	<key>rules</key>
+	<dict>
+		<key>^.*</key>
+		<true/>
+		<key>^.*\.lproj/</key>
+		<dict>
+			<key>optional</key>
+			<true/>
+			<key>weight</key>
+			<real>1000</real>
+		</dict>
+		<key>^.*\.lproj/locversion.plist$</key>
+		<dict>
+			<key>omit</key>
+			<true/>
+			<key>weight</key>
+			<real>1100</real>
+		</dict>
+		<key>^Base\.lproj/</key>
+		<dict>
+			<key>weight</key>
+			<real>1010</real>
+		</dict>
+		<key>^version.plist$</key>
+		<true/>
+	</dict>
+	<key>rules2</key>
+	<dict>
+		<key>.*\.dSYM($|/)</key>
+		<dict>
+			<key>weight</key>
+			<real>11</real>
+		</dict>
+		<key>^(.*/)?\.DS_Store$</key>
+		<dict>
+			<key>omit</key>
+			<true/>
+			<key>weight</key>
+			<real>2000</real>
+		</dict>
+		<key>^.*</key>
+		<true/>
+		<key>^.*\.lproj/</key>
+		<dict>
+			<key>optional</key>
+			<true/>
+			<key>weight</key>
+			<real>1000</real>
+		</dict>
+		<key>^.*\.lproj/locversion.plist$</key>
+		<dict>
+			<key>omit</key>
+			<true/>
+			<key>weight</key>
+			<real>1100</real>
+		</dict>
+		<key>^Base\.lproj/</key>
+		<dict>
+			<key>weight</key>
+			<real>1010</real>
+		</dict>
+		<key>^Info\.plist$</key>
+		<dict>
+			<key>omit</key>
+			<true/>
+			<key>weight</key>
+			<real>20</real>
+		</dict>
+		<key>^PkgInfo$</key>
+		<dict>
+			<key>omit</key>
+			<true/>
+			<key>weight</key>
+			<real>20</real>
+		</dict>
+		<key>^embedded\.provisionprofile$</key>
+		<dict>
+			<key>weight</key>
+			<real>20</real>
+		</dict>
+		<key>^version\.plist$</key>
+		<dict>
+			<key>weight</key>
+			<real>20</real>
+		</dict>
+	</dict>
+</dict>
+</plist>

package/lib/commonjs/NativeCactus.js

@@ -0,0 +1,9 @@
+"use strict";
+
+Object.defineProperty(exports, "__esModule", {
+  value: true
+});
+exports.default = void 0;
+var _reactNative = require("react-native");
+var _default = exports.default = _reactNative.TurboModuleRegistry.get('Cactus');
+//# sourceMappingURL=NativeCactus.js.map

package/lib/commonjs/NativeCactus.js.map

@@ -0,0 +1 @@
+{"version":3,"names":["_reactNative","require","_default","exports","default","TurboModuleRegistry","get"],"sourceRoot":"../../src","sources":["NativeCactus.ts"],"mappings":";;;;;;AACA,IAAAA,YAAA,GAAAC,OAAA;AAAkD,IAAAC,QAAA,GAAAC,OAAA,CAAAC,OAAA,GAganCC,gCAAmB,CAACC,GAAG,CAAO,QAAQ,CAAC","ignoreList":[]}

package/lib/commonjs/chat.js

@@ -0,0 +1,37 @@
+"use strict";
+
+Object.defineProperty(exports, "__esModule", {
+  value: true
+});
+exports.formatChat = formatChat;
+function formatChat(messages) {
+  const chat = [];
+  messages.forEach(currMsg => {
+    const role = currMsg.role || '';
+    let content = '';
+    if ('content' in currMsg) {
+      if (typeof currMsg.content === 'string') {
+        ;
+        ({
+          content
+        } = currMsg);
+      } else if (Array.isArray(currMsg.content)) {
+        currMsg.content.forEach(part => {
+          if ('text' in part) {
+            content += `${content ? '\n' : ''}${part.text}`;
+          }
+        });
+      } else {
+        throw new TypeError("Invalid 'content' type (ref: https://github.com/ggerganov/llama.cpp/issues/8367)");
+      }
+    } else {
+      throw new Error("Missing 'content' (ref: https://github.com/ggerganov/llama.cpp/issues/8367)");
+    }
+    chat.push({
+      role,
+      content
+    });
+  });
+  return chat;
+}
+//# sourceMappingURL=chat.js.map

package/lib/commonjs/chat.js.map

@@ -0,0 +1 @@
+{"version":3,"names":["formatChat","messages","chat","forEach","currMsg","role","content","Array","isArray","part","text","TypeError","Error","push"],"sourceRoot":"../../src","sources":["chat.ts"],"mappings":";;;;;;AAWO,SAASA,UAAUA,CACxBC,QAAsC,EACZ;EAC1B,MAAMC,IAA8B,GAAG,EAAE;EAEzCD,QAAQ,CAACE,OAAO,CAAEC,OAAO,IAAK;IAC5B,MAAMC,IAAY,GAAGD,OAAO,CAACC,IAAI,IAAI,EAAE;IAEvC,IAAIC,OAAe,GAAG,EAAE;IACxB,IAAI,SAAS,IAAIF,OAAO,EAAE;MACxB,IAAI,OAAOA,OAAO,CAACE,OAAO,KAAK,QAAQ,EAAE;QACvC;QAAC,CAAC;UAAEA;QAAQ,CAAC,GAAGF,OAAO;MACzB,CAAC,MAAM,IAAIG,KAAK,CAACC,OAAO,CAACJ,OAAO,CAACE,OAAO,CAAC,EAAE;QACzCF,OAAO,CAACE,OAAO,CAACH,OAAO,CAAEM,IAAI,IAAK;UAChC,IAAI,MAAM,IAAIA,IAAI,EAAE;YAClBH,OAAO,IAAI,GAAGA,OAAO,GAAG,IAAI,GAAG,EAAE,GAAGG,IAAI,CAACC,IAAI,EAAE;UACjD;QACF,CAAC,CAAC;MACJ,CAAC,MAAM;QACL,MAAM,IAAIC,SAAS,CACjB,kFACF,CAAC;MACH;IACF,CAAC,MAAM;MACL,MAAM,IAAIC,KAAK,CACb,6EACF,CAAC;IACH;IAEAV,IAAI,CAACW,IAAI,CAAC;MAAER,IAAI;MAAEC;IAAQ,CAAC,CAAC;EAC9B,CAAC,CAAC;EACF,OAAOJ,IAAI;AACb","ignoreList":[]}