rixty/KoboldAI-Client
1
1
Fork 0
mirror of https://github.com/KoboldAI/KoboldAI-Client.git synced 2025-06-05 21:59:24 +02:00
Code Issues Projects Releases Wiki Activity
3,752 Commits 1 Branch 5 Tags
3adc67c7a498f3a38132658282875f1b95ffed1c
Commit Graph

3752 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
ebolam
23eac76681 Reversing commit 70a25ed6db as that's in one-some's pull 2023-01-10 20:22:52 -05:00
ebolam
bd72b05e12 Fix for editting an action with a matched world info 2023-01-10 20:15:26 -05:00
somebody
ddaf224f83 Rich text STOP NOW!!!! 2023-01-10 18:46:52 -06:00
ebolam
489c1ffd80 Correction to UI2 UI to redirect to API when --no_ui is present. 2023-01-10 16:06:21 -05:00
henk717
4a88e41d14 Pygmalion 6B 2023-01-10 17:22:03 +01:00
ebolam
5d8485273a Remove really old debug message for chat mode 2023-01-10 08:58:09 -05:00
ebolam
9b1138bafa Added in alternative rep pen calculation (log instead of linear application) as an option. 2023-01-10 08:45:55 -05:00
ebolam
70a25ed6db Add in fix for copy/paste of game text pulling in html span elements causing sync errors. 2023-01-10 08:03:43 -05:00
Llama
b10898db51 Merge pull request #11 from henk717/united 
Merge united
 2023-01-09 22:55:14 -08:00
ebolam
33dc4aa749 Remove Debugging 2023-01-09 20:04:05 -05:00
ebolam
4668bc8ec0 Fix for bug that crashes generation when using soft prompts 2023-01-09 19:56:49 -05:00
henk717
1628b789d1 Add Pygmalion 2023-01-09 23:36:43 +01:00
ebolam
3b551ec7ab Added custom model path for model selection. Shows the custom model input box for the user to enter a path and/or a hugging face model 2023-01-09 13:02:57 -05:00
henk717
86f1694290 Merge pull request #254 from one-some/united 
Fix image retry
 2023-01-09 01:34:13 +01:00
somebody
ac5c001fc3 Fix image retry 2023-01-08 18:27:23 -06:00
henk717
535b9bdfdf Merge pull request #251 from one-some/bias-improvements 
Bias improvements
 2023-01-08 17:45:41 +01:00
henk717
419e7c5404 Merge pull request #250 from ebolam/UI2 
Fix for Javascript error on playing last action's TTS audio
 2023-01-08 17:38:42 +01:00
henk717
e45d9a27ae Merge branch 'KoboldAI:main' into united 2023-01-08 17:11:51 +01:00
Henk
857476ef6b ROCm torch version pin 2023-01-08 17:10:59 +01:00
Henk
e2b82b9e17 ROCm Dependency Fixes 2023-01-08 15:31:12 +01:00
somebody
bb3afdf46e Don't allow characters pasted in there! Get out! 2023-01-08 01:11:52 -06:00
somebody
62de50509c UI Fixes 2023-01-08 01:02:17 -06:00
Llama
8d35980911 Merge pull request #10 from henk717/united 
Merge united
 2023-01-07 22:43:37 -08:00
somebody
f6c4bfb390 New bias token control (and lack thereof) 
Allows square bracket syntax for using token ids, curly bracket syntax
for strict phrasing, and normal now biases alternative phrases with
space prefixes
 2023-01-07 23:59:46 -06:00
somebody
76c1398917 Small cleanup with imports 
and unused threading code
 2023-01-07 20:19:31 -06:00
somebody
5f2bcce11a Functional updates (mostly) 2023-01-07 20:11:42 -06:00
ebolam
14b14e26ec Fix for Javascript error on playing last action's TTS audio 2023-01-07 20:30:17 -05:00
Henk
2d85046493 Typo Fix 2023-01-07 21:35:58 +01:00
henk717
e0075d6a34 Merge pull request #249 from ebolam/UI2 
Fix for tortoiseTTS crashing TTS if not installed (should be optional)
 2023-01-07 21:32:56 +01:00
henk717
5f7a5d341b Merge pull request #248 from one-some/hide-load-button-for-now 
Disable load model button on colab/when told to in CLI args
 2023-01-07 21:32:37 +01:00
ebolam
bf7fb52d46 Fix for tortoiseTTS crashing TTS if not installed (should be optional) 2023-01-07 15:22:09 -05:00
henk717
0b1aad7677 Merge pull request #247 from GuiAworld/Themes 
Added the new theme variables to all themes
 2023-01-07 20:42:05 +01:00
Henk
6918a65593 Genre Example is now an example 
You can also just submit your own genre's, so instead of prompting the player to search we use an example term.
 2023-01-07 20:39:49 +01:00
somebody
d298915182 Disable load model button on colab/when prompted 
Currently switching models makes the TPU cry and we do not want to upset
our dearest hardware friend.

In the future it'd probably be best to teach the TPU how to not be
afraid of switching models, provided we have some limitations that
prevent loading itty bitty or way too big models on the TPU.
 2023-01-07 13:38:27 -06:00
henk717
285236bbe5 Merge pull request #246 from GuiAworld/UI2 
A bunch of UI improvements & Material You theme update
 2023-01-07 20:35:53 +01:00
somebody
73f85a1a9e More work on front end 
Its extra shiny
 2023-01-07 00:43:53 -06:00
GuiAworld
982dd115bb Added new theme variables to all themes 2023-01-06 22:58:04 -03:00
GuiAworld
75da540c99 Reverted to the old range sliders 2023-01-06 22:47:18 -03:00
somebody
8e24d91f6d Biasing ui update 2023-01-06 19:21:15 -06:00
GuiAworld
a3e1763969 Merge branch 'united' of https://github.com/henk717/KoboldAI into UI2 2023-01-06 22:01:32 -03:00
GuiAworld
51bf4f5159 Material You theme ported to United 2023-01-06 22:00:55 -03:00
GuiAworld
f3d3e55fec A bunch of UI improvements. 2023-01-06 22:00:33 -03:00
somebody
c6797a59ce Localize variables 
Stop the leaking!
 2023-01-06 18:08:39 -06:00
Henk
0342f0db78 Install submodules on requirements install 
This helps people who clone the repository without knowing submodules need to be cloned, on the dependency installation it will now automatically check the submodules.
 2023-01-06 18:11:48 +01:00
Henk
8fad03431d Merge branch 'main' into united 2023-01-06 16:55:18 +01:00
GuiAworld
075296325f Merge branch 'united' of https://github.com/henk717/KoboldAI into UI2 2023-01-06 12:42:01 -03:00
Henk
7fc5c46c1d Add Safetensors 
Having the dependency adds basic support for safetensor models.
 2023-01-06 16:40:56 +01:00
henk717
b4c2fa0dc1 Merge pull request #245 from pi6am/feat/flexible-memory-length 
Change the maximum memory length to be based on the token budget
 2023-01-06 16:30:58 +01:00
henk717
08e72d0f4d Merge pull request #244 from pi6am/fix/load-non-started 
Fix loading a non-started V1 story json in UI2
 2023-01-06 16:29:52 +01:00
Llama
b28cc01dba Change the maximum memory length to be based on the token budget 
The maximum memory length was hardcoded to 512 tokens independent of
the overall context token budget. This was restrictive when the token
budget was increased, and excessive when the budget was decreased.

Based on feedback, this change modifies the max memory length to be 50%
of the context length. I tested with a memory consisting of the first
five paragraphs of The Three Musketeers, and the change seemed
reasonable with context lengths of 512, 1024 and 2048.

In order to minimize changes, this commit only modifies
max_memory_length; max_prompt_length and max_authornote_length are still
hardcoded as 512 tokens. We might want to make the max prompt length
flexible in the future, but since the author's note is injected into
the story, it's probably fine to restrict its length since a long
author's note will likely derail the AI.
 2023-01-06 00:36:32 -08:00