forked from ggml-org/llama.cpp
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy patharg.cpp
3203 lines (3038 loc) · 135 KB
/
arg.cpp
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
#include "gguf.h" // for reading GGUF splits
#include "arg.h"
#include "common.h"
#include "log.h"
#include "sampling.h"
#include "chat.h"
// fix problem with std::min and std::max
#if defined(_WIN32)
#define WIN32_LEAN_AND_MEAN
#ifndef NOMINMAX
# define NOMINMAX
#endif
#include <windows.h>
#endif
#include <algorithm>
#include <climits>
#include <cstdarg>
#include <filesystem>
#include <fstream>
#include <regex>
#include <set>
#include <string>
#include <thread>
#include <vector>
//#define LLAMA_USE_CURL
#if defined(LLAMA_USE_CURL)
#include <curl/curl.h>
#include <curl/easy.h>
#include <future>
#endif
#include "json-schema-to-grammar.h"
using json = nlohmann::ordered_json;
common_arg & common_arg::set_examples(std::initializer_list<enum llama_example> examples) {
this->examples = std::move(examples);
return *this;
}
common_arg & common_arg::set_excludes(std::initializer_list<enum llama_example> excludes) {
this->excludes = std::move(excludes);
return *this;
}
common_arg & common_arg::set_env(const char * env) {
help = help + "\n(env: " + env + ")";
this->env = env;
return *this;
}
common_arg & common_arg::set_sparam() {
is_sparam = true;
return *this;
}
bool common_arg::in_example(enum llama_example ex) {
return examples.find(ex) != examples.end();
}
bool common_arg::is_exclude(enum llama_example ex) {
return excludes.find(ex) != excludes.end();
}
bool common_arg::get_value_from_env(std::string & output) {
if (env == nullptr) return false;
char * value = std::getenv(env);
if (value) {
output = value;
return true;
}
return false;
}
bool common_arg::has_value_from_env() {
return env != nullptr && std::getenv(env);
}
static std::vector<std::string> break_str_into_lines(std::string input, size_t max_char_per_line) {
std::vector<std::string> result;
std::istringstream iss(input);
std::string line;
auto add_line = [&](const std::string& l) {
if (l.length() <= max_char_per_line) {
result.push_back(l);
} else {
std::istringstream line_stream(l);
std::string word, current_line;
while (line_stream >> word) {
if (current_line.length() + !current_line.empty() + word.length() > max_char_per_line) {
if (!current_line.empty()) result.push_back(current_line);
current_line = word;
} else {
current_line += (!current_line.empty() ? " " : "") + word;
}
}
if (!current_line.empty()) result.push_back(current_line);
}
};
while (std::getline(iss, line)) {
add_line(line);
}
return result;
}
std::string common_arg::to_string() {
// params for printing to console
const static int n_leading_spaces = 40;
const static int n_char_per_line_help = 70; // TODO: detect this based on current console
std::string leading_spaces(n_leading_spaces, ' ');
std::ostringstream ss;
for (const auto arg : args) {
if (arg == args.front()) {
if (args.size() == 1) {
ss << arg;
} else {
// first arg is usually abbreviation, we need padding to make it more beautiful
auto tmp = std::string(arg) + ", ";
auto spaces = std::string(std::max(0, 7 - (int)tmp.size()), ' ');
ss << tmp << spaces;
}
} else {
ss << arg << (arg != args.back() ? ", " : "");
}
}
if (value_hint) ss << " " << value_hint;
if (value_hint_2) ss << " " << value_hint_2;
if (ss.tellp() > n_leading_spaces - 3) {
// current line is too long, add new line
ss << "\n" << leading_spaces;
} else {
// padding between arg and help, same line
ss << std::string(leading_spaces.size() - ss.tellp(), ' ');
}
const auto help_lines = break_str_into_lines(help, n_char_per_line_help);
for (const auto & line : help_lines) {
ss << (&line == &help_lines.front() ? "" : leading_spaces) << line << "\n";
}
return ss.str();
}
//
// downloader
//
struct common_hf_file_res {
std::string repo; // repo name with ":tag" removed
std::string ggufFile;
std::string mmprojFile;
};
#ifdef LLAMA_USE_CURL
#ifdef __linux__
#include <linux/limits.h>
#elif defined(_WIN32)
# if !defined(PATH_MAX)
# define PATH_MAX MAX_PATH
# endif
#elif defined(_AIX)
#include <sys/limits.h>
#else
#include <sys/syslimits.h>
#endif
#define LLAMA_CURL_MAX_URL_LENGTH 2084 // Maximum URL Length in Chrome: 2083
//
// CURL utils
//
using curl_ptr = std::unique_ptr<CURL, decltype(&curl_easy_cleanup)>;
// cannot use unique_ptr for curl_slist, because we cannot update without destroying the old one
struct curl_slist_ptr {
struct curl_slist * ptr = nullptr;
~curl_slist_ptr() {
if (ptr) {
curl_slist_free_all(ptr);
}
}
};
#define CURL_MAX_RETRY 3
#define CURL_RETRY_DELAY_SECONDS 2
static bool curl_perform_with_retry(const std::string & url, CURL * curl, int max_attempts, int retry_delay_seconds) {
int remaining_attempts = max_attempts;
while (remaining_attempts > 0) {
LOG_INF("%s: Trying to download from %s (attempt %d of %d)...\n", __func__ , url.c_str(), max_attempts - remaining_attempts + 1, max_attempts);
CURLcode res = curl_easy_perform(curl);
if (res == CURLE_OK) {
return true;
}
int exponential_backoff_delay = std::pow(retry_delay_seconds, max_attempts - remaining_attempts) * 1000;
LOG_WRN("%s: curl_easy_perform() failed: %s, retrying after %d milliseconds...\n", __func__, curl_easy_strerror(res), exponential_backoff_delay);
remaining_attempts--;
std::this_thread::sleep_for(std::chrono::milliseconds(exponential_backoff_delay));
}
LOG_ERR("%s: curl_easy_perform() failed after %d attempts\n", __func__, max_attempts);
return false;
}
// download one single file from remote URL to local path
static bool common_download_file_single(const std::string & url, const std::string & path, const std::string & bearer_token) {
// Initialize libcurl
curl_ptr curl(curl_easy_init(), &curl_easy_cleanup);
curl_slist_ptr http_headers;
if (!curl) {
LOG_ERR("%s: error initializing libcurl\n", __func__);
return false;
}
bool force_download = false;
// Set the URL, allow to follow http redirection
curl_easy_setopt(curl.get(), CURLOPT_URL, url.c_str());
curl_easy_setopt(curl.get(), CURLOPT_FOLLOWLOCATION, 1L);
http_headers.ptr = curl_slist_append(http_headers.ptr, "User-Agent: llama-cpp");
// Check if hf-token or bearer-token was specified
if (!bearer_token.empty()) {
std::string auth_header = "Authorization: Bearer " + bearer_token;
http_headers.ptr = curl_slist_append(http_headers.ptr, auth_header.c_str());
}
curl_easy_setopt(curl.get(), CURLOPT_HTTPHEADER, http_headers.ptr);
#if defined(_WIN32)
// CURLSSLOPT_NATIVE_CA tells libcurl to use standard certificate store of
// operating system. Currently implemented under MS-Windows.
curl_easy_setopt(curl.get(), CURLOPT_SSL_OPTIONS, CURLSSLOPT_NATIVE_CA);
#endif
// Check if the file already exists locally
auto file_exists = std::filesystem::exists(path);
// If the file exists, check its JSON metadata companion file.
std::string metadata_path = path + ".json";
nlohmann::json metadata;
std::string etag;
std::string last_modified;
if (file_exists) {
// Try and read the JSON metadata file (note: stream autoclosed upon exiting this block).
std::ifstream metadata_in(metadata_path);
if (metadata_in.good()) {
try {
metadata_in >> metadata;
LOG_INF("%s: previous metadata file found %s: %s\n", __func__, metadata_path.c_str(), metadata.dump().c_str());
if (metadata.contains("url") && metadata.at("url").is_string()) {
auto previous_url = metadata.at("url").get<std::string>();
if (previous_url != url) {
LOG_ERR("%s: Model URL mismatch: %s != %s\n", __func__, url.c_str(), previous_url.c_str());
return false;
}
}
if (metadata.contains("etag") && metadata.at("etag").is_string()) {
etag = metadata.at("etag");
}
if (metadata.contains("lastModified") && metadata.at("lastModified").is_string()) {
last_modified = metadata.at("lastModified");
}
} catch (const nlohmann::json::exception & e) {
LOG_ERR("%s: error reading metadata file %s: %s\n", __func__, metadata_path.c_str(), e.what());
return false;
}
}
} else {
LOG_INF("%s: no previous model file found %s\n", __func__, path.c_str());
}
// Send a HEAD request to retrieve the etag and last-modified headers
struct common_load_model_from_url_headers {
std::string etag;
std::string last_modified;
};
common_load_model_from_url_headers headers;
{
typedef size_t(*CURLOPT_HEADERFUNCTION_PTR)(char *, size_t, size_t, void *);
auto header_callback = [](char * buffer, size_t /*size*/, size_t n_items, void * userdata) -> size_t {
common_load_model_from_url_headers * headers = (common_load_model_from_url_headers *) userdata;
static std::regex header_regex("([^:]+): (.*)\r\n");
static std::regex etag_regex("ETag", std::regex_constants::icase);
static std::regex last_modified_regex("Last-Modified", std::regex_constants::icase);
std::string header(buffer, n_items);
std::smatch match;
if (std::regex_match(header, match, header_regex)) {
const std::string & key = match[1];
const std::string & value = match[2];
if (std::regex_match(key, match, etag_regex)) {
headers->etag = value;
} else if (std::regex_match(key, match, last_modified_regex)) {
headers->last_modified = value;
}
}
return n_items;
};
curl_easy_setopt(curl.get(), CURLOPT_NOBODY, 1L); // will trigger the HEAD verb
curl_easy_setopt(curl.get(), CURLOPT_NOPROGRESS, 1L); // hide head request progress
curl_easy_setopt(curl.get(), CURLOPT_HEADERFUNCTION, static_cast<CURLOPT_HEADERFUNCTION_PTR>(header_callback));
curl_easy_setopt(curl.get(), CURLOPT_HEADERDATA, &headers);
bool was_perform_successful = curl_perform_with_retry(url, curl.get(), CURL_MAX_RETRY, CURL_RETRY_DELAY_SECONDS);
if (!was_perform_successful) {
return false;
}
long http_code = 0;
curl_easy_getinfo(curl.get(), CURLINFO_RESPONSE_CODE, &http_code);
if (http_code != 200) {
// HEAD not supported, we don't know if the file has changed
// force trigger downloading
force_download = true;
LOG_ERR("%s: HEAD invalid http status code received: %ld\n", __func__, http_code);
}
}
bool should_download = !file_exists || force_download;
if (!should_download) {
if (!etag.empty() && etag != headers.etag) {
LOG_WRN("%s: ETag header is different (%s != %s): triggering a new download\n", __func__, etag.c_str(), headers.etag.c_str());
should_download = true;
} else if (!last_modified.empty() && last_modified != headers.last_modified) {
LOG_WRN("%s: Last-Modified header is different (%s != %s): triggering a new download\n", __func__, last_modified.c_str(), headers.last_modified.c_str());
should_download = true;
}
}
if (should_download) {
std::string path_temporary = path + ".downloadInProgress";
if (file_exists) {
LOG_WRN("%s: deleting previous downloaded file: %s\n", __func__, path.c_str());
if (remove(path.c_str()) != 0) {
LOG_ERR("%s: unable to delete file: %s\n", __func__, path.c_str());
return false;
}
}
// Set the output file
struct FILE_deleter {
void operator()(FILE * f) const {
fclose(f);
}
};
std::unique_ptr<FILE, FILE_deleter> outfile(fopen(path_temporary.c_str(), "wb"));
if (!outfile) {
LOG_ERR("%s: error opening local file for writing: %s\n", __func__, path.c_str());
return false;
}
typedef size_t(*CURLOPT_WRITEFUNCTION_PTR)(void * data, size_t size, size_t nmemb, void * fd);
auto write_callback = [](void * data, size_t size, size_t nmemb, void * fd) -> size_t {
return fwrite(data, size, nmemb, (FILE *)fd);
};
curl_easy_setopt(curl.get(), CURLOPT_NOBODY, 0L);
curl_easy_setopt(curl.get(), CURLOPT_WRITEFUNCTION, static_cast<CURLOPT_WRITEFUNCTION_PTR>(write_callback));
curl_easy_setopt(curl.get(), CURLOPT_WRITEDATA, outfile.get());
// display download progress
curl_easy_setopt(curl.get(), CURLOPT_NOPROGRESS, 0L);
// helper function to hide password in URL
auto llama_download_hide_password_in_url = [](const std::string & url) -> std::string {
std::size_t protocol_pos = url.find("://");
if (protocol_pos == std::string::npos) {
return url; // Malformed URL
}
std::size_t at_pos = url.find('@', protocol_pos + 3);
if (at_pos == std::string::npos) {
return url; // No password in URL
}
return url.substr(0, protocol_pos + 3) + "********" + url.substr(at_pos);
};
// start the download
LOG_INF("%s: trying to download model from %s to %s (server_etag:%s, server_last_modified:%s)...\n", __func__,
llama_download_hide_password_in_url(url).c_str(), path.c_str(), headers.etag.c_str(), headers.last_modified.c_str());
bool was_perform_successful = curl_perform_with_retry(url, curl.get(), CURL_MAX_RETRY, CURL_RETRY_DELAY_SECONDS);
if (!was_perform_successful) {
return false;
}
long http_code = 0;
curl_easy_getinfo (curl.get(), CURLINFO_RESPONSE_CODE, &http_code);
if (http_code < 200 || http_code >= 400) {
LOG_ERR("%s: invalid http status code received: %ld\n", __func__, http_code);
return false;
}
// Causes file to be closed explicitly here before we rename it.
outfile.reset();
// Write the updated JSON metadata file.
metadata.update({
{"url", url},
{"etag", headers.etag},
{"lastModified", headers.last_modified}
});
std::ofstream(metadata_path) << metadata.dump(4);
LOG_INF("%s: file metadata saved: %s\n", __func__, metadata_path.c_str());
if (rename(path_temporary.c_str(), path.c_str()) != 0) {
LOG_ERR("%s: unable to rename file: %s to %s\n", __func__, path_temporary.c_str(), path.c_str());
return false;
}
}
return true;
}
// download multiple files from remote URLs to local paths
// the input is a vector of pairs <url, path>
static bool common_download_file_multiple(const std::vector<std::pair<std::string, std::string>> & urls, const std::string & bearer_token) {
// Prepare download in parallel
std::vector<std::future<bool>> futures_download;
for (auto const & item : urls) {
futures_download.push_back(std::async(std::launch::async, [bearer_token](const std::pair<std::string, std::string> & it) -> bool {
return common_download_file_single(it.first, it.second, bearer_token);
}, item));
}
// Wait for all downloads to complete
for (auto & f : futures_download) {
if (!f.get()) {
return false;
}
}
return true;
}
static bool common_download_model(
const common_params_model & model,
const std::string & bearer_token) {
// Basic validation of the model.url
if (model.url.empty()) {
LOG_ERR("%s: invalid model url\n", __func__);
return false;
}
if (!common_download_file_single(model.url, model.path, bearer_token)) {
return false;
}
// check for additional GGUFs split to download
int n_split = 0;
{
struct gguf_init_params gguf_params = {
/*.no_alloc = */ true,
/*.ctx = */ NULL,
};
auto * ctx_gguf = gguf_init_from_file(model.path.c_str(), gguf_params);
if (!ctx_gguf) {
LOG_ERR("\n%s: failed to load input GGUF from %s\n", __func__, model.path.c_str());
return false;
}
auto key_n_split = gguf_find_key(ctx_gguf, LLM_KV_SPLIT_COUNT);
if (key_n_split >= 0) {
n_split = gguf_get_val_u16(ctx_gguf, key_n_split);
}
gguf_free(ctx_gguf);
}
if (n_split > 1) {
char split_prefix[PATH_MAX] = {0};
char split_url_prefix[LLAMA_CURL_MAX_URL_LENGTH] = {0};
// Verify the first split file format
// and extract split URL and PATH prefixes
{
if (!llama_split_prefix(split_prefix, sizeof(split_prefix), model.path.c_str(), 0, n_split)) {
LOG_ERR("\n%s: unexpected model file name: %s n_split=%d\n", __func__, model.path.c_str(), n_split);
return false;
}
if (!llama_split_prefix(split_url_prefix, sizeof(split_url_prefix), model.url.c_str(), 0, n_split)) {
LOG_ERR("\n%s: unexpected model url: %s n_split=%d\n", __func__, model.url.c_str(), n_split);
return false;
}
}
std::vector<std::pair<std::string, std::string>> urls;
for (int idx = 1; idx < n_split; idx++) {
char split_path[PATH_MAX] = {0};
llama_split_path(split_path, sizeof(split_path), split_prefix, idx, n_split);
char split_url[LLAMA_CURL_MAX_URL_LENGTH] = {0};
llama_split_path(split_url, sizeof(split_url), split_url_prefix, idx, n_split);
if (std::string(split_path) == model.path) {
continue; // skip the already downloaded file
}
urls.push_back({split_url, split_path});
}
// Download in parallel
common_download_file_multiple(urls, bearer_token);
}
return true;
}
/**
* Allow getting the HF file from the HF repo with tag (like ollama), for example:
* - bartowski/Llama-3.2-3B-Instruct-GGUF:q4
* - bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M
* - bartowski/Llama-3.2-3B-Instruct-GGUF:q5_k_s
* Tag is optional, default to "latest" (meaning it checks for Q4_K_M first, then Q4, then if not found, return the first GGUF file in repo)
*
* Return pair of <repo, file> (with "repo" already having tag removed)
*
* Note: we use the Ollama-compatible HF API, but not using the blobId. Instead, we use the special "ggufFile" field which returns the value for "hf_file". This is done to be backward-compatible with existing cache files.
*/
static struct common_hf_file_res common_get_hf_file(const std::string & hf_repo_with_tag, const std::string & bearer_token) {
auto parts = string_split<std::string>(hf_repo_with_tag, ':');
std::string tag = parts.size() > 1 ? parts.back() : "latest";
std::string hf_repo = parts[0];
if (string_split<std::string>(hf_repo, '/').size() != 2) {
throw std::invalid_argument("error: invalid HF repo format, expected <user>/<model>[:quant]\n");
}
// fetch model info from Hugging Face Hub API
curl_ptr curl(curl_easy_init(), &curl_easy_cleanup);
curl_slist_ptr http_headers;
std::string res_str;
std::string model_endpoint = get_model_endpoint();
std::string url = model_endpoint + "v2/" + hf_repo + "/manifests/" + tag;
curl_easy_setopt(curl.get(), CURLOPT_URL, url.c_str());
curl_easy_setopt(curl.get(), CURLOPT_NOPROGRESS, 1L);
typedef size_t(*CURLOPT_WRITEFUNCTION_PTR)(void * ptr, size_t size, size_t nmemb, void * data);
auto write_callback = [](void * ptr, size_t size, size_t nmemb, void * data) -> size_t {
static_cast<std::string *>(data)->append((char * ) ptr, size * nmemb);
return size * nmemb;
};
curl_easy_setopt(curl.get(), CURLOPT_WRITEFUNCTION, static_cast<CURLOPT_WRITEFUNCTION_PTR>(write_callback));
curl_easy_setopt(curl.get(), CURLOPT_WRITEDATA, &res_str);
#if defined(_WIN32)
curl_easy_setopt(curl.get(), CURLOPT_SSL_OPTIONS, CURLSSLOPT_NATIVE_CA);
#endif
if (!bearer_token.empty()) {
std::string auth_header = "Authorization: Bearer " + bearer_token;
http_headers.ptr = curl_slist_append(http_headers.ptr, auth_header.c_str());
}
// Important: the User-Agent must be "llama-cpp" to get the "ggufFile" field in the response
http_headers.ptr = curl_slist_append(http_headers.ptr, "User-Agent: llama-cpp");
http_headers.ptr = curl_slist_append(http_headers.ptr, "Accept: application/json");
curl_easy_setopt(curl.get(), CURLOPT_HTTPHEADER, http_headers.ptr);
CURLcode res = curl_easy_perform(curl.get());
if (res != CURLE_OK) {
throw std::runtime_error("error: cannot make GET request to HF API");
}
long res_code;
std::string ggufFile = "";
std::string mmprojFile = "";
curl_easy_getinfo(curl.get(), CURLINFO_RESPONSE_CODE, &res_code);
if (res_code == 200) {
// extract ggufFile.rfilename in json, using regex
{
std::regex pattern("\"ggufFile\"[\\s\\S]*?\"rfilename\"\\s*:\\s*\"([^\"]+)\"");
std::smatch match;
if (std::regex_search(res_str, match, pattern)) {
ggufFile = match[1].str();
}
}
// extract mmprojFile.rfilename in json, using regex
{
std::regex pattern("\"mmprojFile\"[\\s\\S]*?\"rfilename\"\\s*:\\s*\"([^\"]+)\"");
std::smatch match;
if (std::regex_search(res_str, match, pattern)) {
mmprojFile = match[1].str();
}
}
} else if (res_code == 401) {
throw std::runtime_error("error: model is private or does not exist; if you are accessing a gated model, please provide a valid HF token");
} else {
throw std::runtime_error(string_format("error from HF API, response code: %ld, data: %s", res_code, res_str.c_str()));
}
// check response
if (ggufFile.empty()) {
throw std::runtime_error("error: model does not have ggufFile");
}
return { hf_repo, ggufFile, mmprojFile };
}
#else
static bool common_download_file_single(const std::string &, const std::string &, const std::string &) {
LOG_ERR("error: built without CURL, cannot download model from internet\n");
return false;
}
static bool common_download_file_multiple(const std::vector<std::pair<std::string, std::string>> &, const std::string &) {
LOG_ERR("error: built without CURL, cannot download model from the internet\n");
return false;
}
static bool common_download_model(
const common_params_model &,
const std::string &) {
LOG_ERR("error: built without CURL, cannot download model from the internet\n");
return false;
}
static struct common_hf_file_res common_get_hf_file(const std::string &, const std::string &) {
LOG_ERR("error: built without CURL, cannot download model from the internet\n");
return {};
}
#endif // LLAMA_USE_CURL
//
// utils
//
static void common_params_handle_model(
struct common_params_model & model,
const std::string & bearer_token,
const std::string & model_path_default,
bool is_mmproj = false) { // TODO: move is_mmproj to an enum when we have more files?
// handle pre-fill default model path and url based on hf_repo and hf_file
{
if (!model.hf_repo.empty()) {
// short-hand to avoid specifying --hf-file -> default it to --model
if (model.hf_file.empty()) {
if (model.path.empty()) {
auto auto_detected = common_get_hf_file(model.hf_repo, bearer_token);
if (auto_detected.repo.empty() || auto_detected.ggufFile.empty()) {
exit(1); // built without CURL, error message already printed
}
model.hf_repo = auto_detected.repo;
model.hf_file = is_mmproj ? auto_detected.mmprojFile : auto_detected.ggufFile;
} else {
model.hf_file = model.path;
}
}
std::string model_endpoint = get_model_endpoint();
model.url = model_endpoint + model.hf_repo + "/resolve/main/" + model.hf_file;
// make sure model path is present (for caching purposes)
if (model.path.empty()) {
// this is to avoid different repo having same file name, or same file name in different subdirs
std::string filename = model.hf_repo + "_" + model.hf_file;
// to make sure we don't have any slashes in the filename
string_replace_all(filename, "/", "_");
model.path = fs_get_cache_file(filename);
}
} else if (!model.url.empty()) {
if (model.path.empty()) {
auto f = string_split<std::string>(model.url, '#').front();
f = string_split<std::string>(f, '?').front();
model.path = fs_get_cache_file(string_split<std::string>(f, '/').back());
}
} else if (model.path.empty()) {
model.path = model_path_default;
}
}
// then, download it if needed
if (!model.url.empty()) {
bool ok = common_download_model(model, bearer_token);
if (!ok) {
LOG_ERR("error: failed to download model from %s\n", model.url.c_str());
exit(1);
}
}
}
const std::vector<ggml_type> kv_cache_types = {
GGML_TYPE_F32,
GGML_TYPE_F16,
GGML_TYPE_BF16,
GGML_TYPE_Q8_0,
GGML_TYPE_Q4_0,
GGML_TYPE_Q4_1,
GGML_TYPE_IQ4_NL,
GGML_TYPE_Q5_0,
GGML_TYPE_Q5_1,
};
static ggml_type kv_cache_type_from_str(const std::string & s) {
for (const auto & type : kv_cache_types) {
if (ggml_type_name(type) == s) {
return type;
}
}
throw std::runtime_error("Unsupported cache type: " + s);
}
static std::string get_all_kv_cache_types() {
std::ostringstream msg;
for (const auto & type : kv_cache_types) {
msg << ggml_type_name(type) << (&type == &kv_cache_types.back() ? "" : ", ");
}
return msg.str();
}
//
// CLI argument parsing functions
//
static bool common_params_parse_ex(int argc, char ** argv, common_params_context & ctx_arg) {
std::string arg;
const std::string arg_prefix = "--";
common_params & params = ctx_arg.params;
std::unordered_map<std::string, common_arg *> arg_to_options;
for (auto & opt : ctx_arg.options) {
for (const auto & arg : opt.args) {
arg_to_options[arg] = &opt;
}
}
// handle environment variables
for (auto & opt : ctx_arg.options) {
std::string value;
if (opt.get_value_from_env(value)) {
try {
if (opt.handler_void && (value == "1" || value == "true")) {
opt.handler_void(params);
}
if (opt.handler_int) {
opt.handler_int(params, std::stoi(value));
}
if (opt.handler_string) {
opt.handler_string(params, value);
continue;
}
} catch (std::exception & e) {
throw std::invalid_argument(string_format(
"error while handling environment variable \"%s\": %s\n\n", opt.env, e.what()));
}
}
}
// handle command line arguments
auto check_arg = [&](int i) {
if (i+1 >= argc) {
throw std::invalid_argument("expected value for argument");
}
};
for (int i = 1; i < argc; i++) {
const std::string arg_prefix = "--";
std::string arg = argv[i];
if (arg.compare(0, arg_prefix.size(), arg_prefix) == 0) {
std::replace(arg.begin(), arg.end(), '_', '-');
}
if (arg_to_options.find(arg) == arg_to_options.end()) {
throw std::invalid_argument(string_format("error: invalid argument: %s", arg.c_str()));
}
auto opt = *arg_to_options[arg];
if (opt.has_value_from_env()) {
fprintf(stderr, "warn: %s environment variable is set, but will be overwritten by command line argument %s\n", opt.env, arg.c_str());
}
try {
if (opt.handler_void) {
opt.handler_void(params);
continue;
}
// arg with single value
check_arg(i);
std::string val = argv[++i];
if (opt.handler_int) {
opt.handler_int(params, std::stoi(val));
continue;
}
if (opt.handler_string) {
opt.handler_string(params, val);
continue;
}
// arg with 2 values
check_arg(i);
std::string val2 = argv[++i];
if (opt.handler_str_str) {
opt.handler_str_str(params, val, val2);
continue;
}
} catch (std::exception & e) {
throw std::invalid_argument(string_format(
"error while handling argument \"%s\": %s\n\n"
"usage:\n%s\n\nto show complete usage, run with -h",
arg.c_str(), e.what(), arg_to_options[arg]->to_string().c_str()));
}
}
postprocess_cpu_params(params.cpuparams, nullptr);
postprocess_cpu_params(params.cpuparams_batch, ¶ms.cpuparams);
postprocess_cpu_params(params.speculative.cpuparams, ¶ms.cpuparams);
postprocess_cpu_params(params.speculative.cpuparams_batch, ¶ms.cpuparams_batch);
if (params.prompt_cache_all && (params.interactive || params.interactive_first)) {
throw std::invalid_argument("error: --prompt-cache-all not supported in interactive mode yet\n");
}
common_params_handle_model(params.model, params.hf_token, DEFAULT_MODEL_PATH);
common_params_handle_model(params.speculative.model, params.hf_token, "");
common_params_handle_model(params.vocoder.model, params.hf_token, "");
// allow --mmproj to be set from -hf
// assuming that mmproj is always in the same repo as text model
if (!params.model.hf_repo.empty() && ctx_arg.ex == LLAMA_EXAMPLE_LLAVA) {
params.mmproj.hf_repo = params.model.hf_repo;
}
common_params_handle_model(params.mmproj, params.hf_token, "", true);
if (params.escape) {
string_process_escapes(params.prompt);
string_process_escapes(params.input_prefix);
string_process_escapes(params.input_suffix);
for (auto & antiprompt : params.antiprompt) {
string_process_escapes(antiprompt);
}
for (auto & seq_breaker : params.sampling.dry_sequence_breakers) {
string_process_escapes(seq_breaker);
}
}
if (!params.kv_overrides.empty()) {
params.kv_overrides.emplace_back();
params.kv_overrides.back().key[0] = 0;
}
if (!params.tensor_buft_overrides.empty()) {
params.tensor_buft_overrides.push_back({nullptr, nullptr});
}
if (params.reranking && params.embedding) {
throw std::invalid_argument("error: either --embedding or --reranking can be specified, but not both");
}
if (!params.chat_template.empty() && !common_chat_verify_template(params.chat_template, params.use_jinja)) {
throw std::runtime_error(string_format(
"error: the supplied chat template is not supported: %s%s\n",
params.chat_template.c_str(),
params.use_jinja ? "" : "\nnote: llama.cpp was started without --jinja, we only support commonly used templates"
));
}
return true;
}
static void common_params_print_usage(common_params_context & ctx_arg) {
auto print_options = [](std::vector<common_arg *> & options) {
for (common_arg * opt : options) {
printf("%s", opt->to_string().c_str());
}
};
std::vector<common_arg *> common_options;
std::vector<common_arg *> sparam_options;
std::vector<common_arg *> specific_options;
for (auto & opt : ctx_arg.options) {
// in case multiple LLAMA_EXAMPLE_* are set, we prioritize the LLAMA_EXAMPLE_* matching current example
if (opt.is_sparam) {
sparam_options.push_back(&opt);
} else if (opt.in_example(ctx_arg.ex)) {
specific_options.push_back(&opt);
} else {
common_options.push_back(&opt);
}
}
printf("----- common params -----\n\n");
print_options(common_options);
printf("\n\n----- sampling params -----\n\n");
print_options(sparam_options);
// TODO: maybe convert enum llama_example to string
printf("\n\n----- example-specific params -----\n\n");
print_options(specific_options);
}
static void common_params_print_completion(common_params_context & ctx_arg) {
std::vector<common_arg *> common_options;
std::vector<common_arg *> sparam_options;
std::vector<common_arg *> specific_options;
for (auto & opt : ctx_arg.options) {
if (opt.is_sparam) {
sparam_options.push_back(&opt);
} else if (opt.in_example(ctx_arg.ex)) {
specific_options.push_back(&opt);
} else {
common_options.push_back(&opt);
}
}
printf("_llama_completions() {\n");
printf(" local cur prev opts\n");
printf(" COMPREPLY=()\n");
printf(" cur=\"${COMP_WORDS[COMP_CWORD]}\"\n");
printf(" prev=\"${COMP_WORDS[COMP_CWORD-1]}\"\n\n");
printf(" opts=\"");
auto print_options = [](const std::vector<common_arg *> & options) {
for (const common_arg * opt : options) {
for (const char * arg : opt->args) {
printf("%s ", arg);
}
}
};
print_options(common_options);
print_options(sparam_options);
print_options(specific_options);
printf("\"\n\n");
printf(" case \"$prev\" in\n");
printf(" --model)\n");
printf(" COMPREPLY=( $(compgen -f -X '!*.gguf' -- \"$cur\") $(compgen -d -- \"$cur\") )\n");
printf(" return 0\n");
printf(" ;;\n");
printf(" --grammar-file)\n");
printf(" COMPREPLY=( $(compgen -f -X '!*.gbnf' -- \"$cur\") $(compgen -d -- \"$cur\") )\n");
printf(" return 0\n");
printf(" ;;\n");
printf(" --chat-template-file)\n");
printf(" COMPREPLY=( $(compgen -f -X '!*.jinja' -- \"$cur\") $(compgen -d -- \"$cur\") )\n");
printf(" return 0\n");
printf(" ;;\n");
printf(" *)\n");
printf(" COMPREPLY=( $(compgen -W \"${opts}\" -- \"$cur\") )\n");
printf(" return 0\n");
printf(" ;;\n");
printf(" esac\n");
printf("}\n\n");
std::set<std::string> executables = {
"llama-batched",
"llama-batched-bench",
"llama-bench",
"llama-cli",
"llama-convert-llama2c-to-ggml",
"llama-cvector-generator",
"llama-embedding",
"llama-eval-callback",
"llama-export-lora",
"llama-gbnf-validator",
"llama-gen-docs",
"llama-gguf",
"llama-gguf-hash",
"llama-gguf-split",
"llama-gritlm",
"llama-imatrix",
"llama-infill",
"llama-llava-cli",
"llama-llava-clip-quantize-cli",
"llama-lookahead",
"llama-lookup",
"llama-lookup-create",
"llama-lookup-merge",
"llama-lookup-stats",
"llama-minicpmv-cli",
"llama-parallel",
"llama-passkey",
"llama-perplexity",
"llama-q8dot",
"llama-quantize",
"llama-quantize-stats",
"llama-qwen2vl-cli",
"llama-retrieval",
"llama-run",
"llama-save-load-state",
"llama-server",
"llama-simple",
"llama-simple-chat",
"llama-speculative",