Commit Graph

Select branches

Hide Pull Requests

main

#102

#103

#112

#114

#116

#117

#119

#12

#123

#126

#127

#128

#129

#130

#132

#133

#134

#136

#137

#138

#140

#141

#143

#144

#145

#146

#147

#148

#152

#156

#157

#158

#161

#162

#165

#168

#172

#175

#179

#180

#184

#185

#191

#194

#196

#197

#197

#198

#20

#201

#21

#213

#237

#239

#24

#243

#245

#25

#255

#26

#267

#27

#276

#277

#28

#291

#306

#307

#308

#309

#309

#311

#318

#342

#356

#37

#371

#373

#375

#38

#390

#4

#40

#409

#410

#412

#413

#416

#420

#424

#424

#428

#429

#43

#431

#435

#436

#436

#442

#442

#443

#45

#453

#46

#47

#48

#49

#50

#53

#56

#58

#59

#60

#65

#66

#67

#68

#69

#70

#71

#72

#73

#74

#76

#77

#78

#79

#80

#81

#82

#84

#90

#91

#95

#97

#98

#99

1.18.2

1.19.0

1.19.1

1.19.2

dummy

Mono Color

• a73420c49c really really really sketchy breakmodel implementation somebody 2023-07-24 17:15:59 -05:00
• ec040620ec Merge branch 'united' of https://github.com/henk717/KoboldAI into qptq-lazy somebody 2023-07-24 13:26:16 -05:00
• 34aa333c44 Last debug somebody 2023-07-24 13:11:06 -05:00
• 43a4abaf63 Remove even more debug somebody 2023-07-24 13:10:33 -05:00
• 929917efe9 Remove shrieking somebody 2023-07-24 13:09:43 -05:00
• 4a6cccb002 Import fix somebody 2023-07-24 13:09:15 -05:00
• a6aafb2525 GPTQ: Patch QuantLinear to not use CPU RAM somebody 2023-07-24 13:07:30 -05:00
• 9cc6972c1c Shh! somebody 2023-07-24 11:30:33 -05:00
• 30640acca7 Editor: Don't allow editing or syncing during generation somebody 2023-07-24 11:26:20 -05:00
• fc7fa991d5 Streaming: Fix streaming not being cleaned up before commentator speaks somebody 2023-07-24 10:57:24 -05:00
• 04a5e05692
  Merge pull request #422 from one-some/fix-prioritization henk717 2023-07-24 17:40:36 +02:00
• 2fb877db40 Backends: Probably fix sorting somebody 2023-07-24 10:28:22 -05:00
• 81e4c8a807 Backends: Fix GPTQ priority somebody 2023-07-24 10:25:44 -05:00
• ba313883b6 Merge branch 'united' of https://github.com/henk717/KoboldAI into submit-ctx-menu somebody 2023-07-24 10:09:38 -05:00
• 1df03d9a27 Basic somebody 2023-07-23 20:54:04 -05:00
• 30495cf8d8 Fix GPT2 Henk 2023-07-24 02:05:07 +02:00
• 9fc9cb92f7 Fancy streaming by default Henk 2023-07-24 01:39:16 +02:00
• c60a29e544
  Merge pull request #419 from one-some/streaming-fix-2 henk717 2023-07-24 01:31:51 +02:00
• 70d2da55e5 Readme changes Henk 2023-07-24 01:03:46 +02:00
• 1371150cbd Merge branch 'united' of https://github.com/henk717/KoboldAI into wi-fixes somebody 2023-07-23 18:00:28 -05:00
• f4593ed04b Streaming: Fix bad streamingwindow sync somebody 2023-07-23 17:50:53 -05:00
• 8de610df8c Streaming: Rework single-gen streaming somebody 2023-07-23 17:32:52 -05:00
• a963c97acb Make 4-bit the default part 2 Henk 2023-07-24 00:06:20 +02:00
• 3409853dfc Remove GPTQ for Colab Henk 2023-07-23 23:26:35 +02:00
• 70dddf9fdc Prioritize GPTQ Henk 2023-07-23 23:22:02 +02:00
• ae9ec38ae2
  Merge pull request #418 from one-some/united henk717 2023-07-23 23:09:42 +02:00
• 0f913275a9 4-bit as Default Henk 2023-07-23 23:08:11 +02:00
• 3aa677ce11 Indexed prioritization somebody 2023-07-23 16:04:23 -05:00
• 89637ae9d7 GPTQ Requirements Henk 2023-07-23 22:51:47 +02:00
• 1facc73b66
  Merge pull request #367 from 0cc4m/4bit-plugin henk717 2023-07-23 22:32:20 +02:00
• 73953068c0 Remove exllama backend, pending further fixes 0cc4m 2023-07-23 22:12:31 +02:00
• 973aea12ea Only import big python modules for GPTQ once they get used 0cc4m 2023-07-23 22:07:34 +02:00
• 49740aa5ab Fix ntk alpha 0cc4m 2023-07-23 21:56:48 +02:00
• e33a58b74a Adventure stoppers = regex Henk 2023-07-23 20:45:48 +02:00
• d70481874c
  Merge pull request #334 from YellowRoseCx/YellowRoseCx-advpatch-1 henk717 2023-07-23 20:38:51 +02:00
• 66fb8b8937
  Merge pull request #415 from one-some/fixes-forever henk717 2023-07-23 20:05:38 +02:00
• b3b67bf50d
  Merge pull request #417 from LostRuins/concedo_united henk717 2023-07-23 16:11:23 +02:00
• dd8e5f5d05 updated lite to v50 Concedo 2023-07-23 21:40:08 +08:00
• 31a984aa3d Automatically install exllama module 0cc4m 2023-07-23 07:33:51 +02:00
• a9aa04fd1b Merge remote-tracking branch 'upstream/united' into 4bit-plugin 0cc4m 2023-07-23 07:18:58 +02:00
• 09bb1021dd Fallback to transformers if hf_bleeding_edge not available 0cc4m 2023-07-23 07:14:23 +02:00
• 748e5ef318 Add sliders for exllama context size and related methods 0cc4m 2023-07-23 07:11:28 +02:00
• 3995b3f93b WI: Make delete button pretty somebody 2023-07-22 18:18:21 -05:00
• bd542336f9 WI: Make the noun thingey more intuitive somebody 2023-07-22 18:12:30 -05:00
• 65cf6806a8 WI: Enter to save name somebody 2023-07-22 18:04:35 -05:00
• 132ed1b507 UI: Make experimental tooltip more ominous somebody 2023-07-22 17:53:28 -05:00
• 33cec5cc9c UI: Fix visual inconsistancies in sidebar somebody 2023-07-22 17:45:02 -05:00
• 432418ed1e UI: Possibly more clear tooltips somebody 2023-07-22 17:44:32 -05:00
• ccbfad1a13 UI: Make welcome text links have underlines somebody 2023-07-22 17:35:16 -05:00
• c7b128829c WI: Fix visual oddness with more than one row of tags somebody 2023-07-22 17:30:43 -05:00
• 6b26cbbd0a Backends: Fix ReadOnly somebody 2023-07-22 17:20:40 -05:00
• 68c6030ab0 WI: Don't explode when user uploads image without a save somebody 2023-07-22 17:04:45 -05:00
• cf27d44f62 WI: Tag polish somebody 2023-07-22 16:50:29 -05:00
• 7a5d813b92 Reimplement HF workaround only for llama Henk 2023-07-22 16:59:49 +02:00
• 8dd7b93a6c HF's workaround breaks stuff Henk 2023-07-22 16:29:55 +02:00
• fa9d17b3d3 HF 4.31 Henk 2023-07-22 15:25:14 +02:00
• bc8ba91429 Private Mode improvements somebody 2023-07-21 21:44:10 -05:00
• 7823da564e Link to Lite Henk 2023-07-22 04:04:17 +02:00
• a93c9d20b1 Don't let logo container gobble up clicks somebody 2023-07-21 19:01:51 -05:00
• 79b1ef1aac Fix "hide welcome logo" tweak somebody 2023-07-21 19:01:40 -05:00
• 9188323331 Biases: Don't crash on empty token seq somebody 2023-07-21 18:56:29 -05:00
• 5f4216730e Make logit bias work correctly(?) when prob is -inf somebody 2023-07-21 18:33:35 -05:00
• 418f341560 Fix a/n depth being visually apart from a/n somebody 2023-07-21 18:13:57 -05:00
• 560fb3bd2d Fix occasional action highlight issue somebody 2023-07-21 18:08:21 -05:00
• 83e5c29260
  Merge pull request #413 from one-some/bug-hunt henk717 2023-07-22 00:34:46 +02:00
• e68972a270 Fix WI comments somebody 2023-07-21 16:14:13 -05:00
• 6e7b0794ea Context Menu: Fix for elements with a context-menu attribute but... somebody 2023-07-21 15:40:07 -05:00
• e5d0a597a1 Generation Mode: UNTIL_EOS somebody 2023-07-21 15:36:32 -05:00
• c78401bd12 Fix gen mode on first generation somebody 2023-07-21 15:22:14 -05:00
• 8d5ae38b45 Context Menu: Show if gen mode is supported somebody 2023-07-21 14:29:41 -05:00
• b8671cce09 Context Menu: Change positioning algorithm for y-axis somebody 2023-07-21 13:48:23 -05:00
• 1c4157a41b Maybe another time somebody 2023-07-21 13:33:38 -05:00
• 3a43b254b8 Add basic support for some of the quick stoppers somebody 2023-07-21 13:27:30 -05:00
• a17d7aae60 Easier english Henk 2023-07-21 19:42:49 +02:00
• da9b54ec1c Don't show API link during load Henk 2023-07-21 19:31:38 +02:00
• fa0a099943 Update comment somebody 2023-07-21 10:38:17 -05:00
• 432cdc9a08 Fix models with good pad tokens Henk 2023-07-21 16:39:58 +02:00
• ec745d8b80 Dont accidentally block pad tokens Henk 2023-07-21 16:25:32 +02:00
• 6cf63f781a YEAAAAAAAAAA onesome 2023-07-21 01:58:57 -05:00
• 46c377b0c3 Context Menu: Add stubs for new temporary stoppingcriteria idea onesome 2023-07-21 00:53:48 -05:00
• 4921040fb4 Context Menu: Make things a little less bloaty onesome 2023-07-21 00:52:12 -05:00
• 34a98d2962 Context Menu: Small visual fixes onesome 2023-07-21 00:48:02 -05:00
• 4335d1f46a API: Fix /world_info somebody 2023-07-19 13:18:45 -05:00
• 2d80f2ebb5 API: Fix getstorynums somebody 2023-07-19 13:08:57 -05:00
• 9726d12ede API: Fix /story/end (POST) somebody 2023-07-19 13:05:35 -05:00
• 6da7a9629a API: Fix /story/load somebody 2023-07-19 13:01:07 -05:00
• b9b3cd3aba API: Fix /story somebody 2023-07-19 12:02:53 -05:00
• 813e210127 Bump tiny API version somebody 2023-07-19 11:52:49 -05:00
• fef42a6273 API: Fix loading somebody 2023-07-19 11:52:39 -05:00
• dc4404f29c
  Merge pull request #409 from nkpz/bnb8bit henk717 2023-07-19 14:22:44 +02:00
• 9581e51476 feature(load model): select control for quantization level Nick Perez 2023-07-19 07:58:12 -04:00
• 58908ab846 Revert aiserver.py changes 0cc4m 2023-07-19 07:14:03 +02:00
• 19f511dc9f Load GPTQ module from GPTQ repo docs 0cc4m 2023-07-19 07:12:37 +02:00
• 1c5da2bbf3 Move pip docs from KoboldAI into GPTQ repo 0cc4m 2023-07-19 07:08:39 +02:00
• 7516ecf00d Merge upstream changes, fix conflict 0cc4m 2023-07-19 07:02:29 +02:00
• c84d063be8 Revert settings changes 0cc4m 2023-07-19 07:01:11 +02:00
• 9aa6c5fbbf Merge upstream changes, fix conflict, adapt backends to changes 0cc4m 2023-07-19 06:56:09 +02:00
• 0142913060 8 bit toggle, fix for broken toggle values Nick Perez 2023-07-18 23:29:38 -04:00
• 22e7baec52 Permit CPU layers on 4-bit (Worse than GGML) Henk 2023-07-18 21:44:34 +02:00
• 5f2600d338
  Merge pull request #406 from ebolam/Model_Plugins henk717 2023-07-18 02:42:23 +02:00