4488cee65d
Add search filtering for past chats
2023-08-18 23:45:05 -04:00
3019b889cd
Filter group candidates before mapping
2023-08-19 03:22:10 +03:00
1305f9dc4c
Fix prompt manager sometimes erroring in groups with null members
2023-08-19 03:19:12 +03:00
059647b286
Dynamically update group avatars
2023-08-19 03:13:40 +03:00
f27107e0ef
Refactor and optimize group members display
2023-08-19 02:53:05 +03:00
0a0724255d
fix default context
2023-08-18 23:58:03 +02:00
bcc66de1df
multi-line clean-up message
2023-08-18 23:28:11 +02:00
599a912a5a
deduplicated multi-line stopping strings
2023-08-18 23:02:59 +02:00
602c5cd791
Add pagination for characters list view
2023-08-18 23:13:15 +03:00
26e2b0c8fa
Fix whitespace issues
2023-08-18 21:18:54 +02:00
7e5d08874f
Handle last_output_sequence like output_sequence
2023-08-18 21:18:27 +02:00
a40df0c117
Added simple-proxy-for-tavern context template and instruct preset
2023-08-18 21:16:04 +02:00
6a89967144
Added Roleplay context template and updated instruct preset
2023-08-18 21:15:44 +02:00
0f162c1803
Save Novel preamble to presets
2023-08-18 17:43:50 +03:00
a0ed362591
move char panel token counter to desc box
2023-08-18 22:47:43 +09:00
0c5c8c50ad
half-fix for small boxes in mobile no spoiler view
2023-08-18 22:45:40 +09:00
b645746a99
Merge pull request #963 from deffcolony/staging
...
removes width: 100% from img.expression
2023-08-18 16:20:11 +03:00
33a5a026d5
removes width: 100% from img.expression
...
this will fix the image resolution on mobile
2023-08-18 15:13:41 +02:00
08a0709186
Escape "Match whole words" regex
2023-08-18 14:54:38 +03:00
9bd9ddb9e1
Fix expressions display
2023-08-18 14:47:59 +03:00
02cac8db8c
Disable escaping of HTML/tags in templates
2023-08-18 13:53:54 +03:00
c4d8d70103
Improve phone usability of sortable drags
2023-08-18 13:41:46 +03:00
1e0e5225ca
Fix quiet prompt generations for Novel with streaming enabled.
2023-08-18 12:56:20 +03:00
5528a493a6
Do not insert an extra newline if example separator is empty
2023-08-18 12:32:52 +03:00
75378eff4e
#937 Add dynamic chat backgrounds generation
2023-08-18 00:41:21 +03:00
c24f636a10
Add Last output sequence setting for instruct mode
2023-08-17 23:33:27 +03:00
e147ad1e18
Merge pull request #958 from Tony-sama/staging
...
Coqui api full model available (Experimental)
2023-08-17 22:55:00 +03:00
80092b3170
#790 Simplify local prompt formatting. Use handlebars to render story string.
2023-08-17 22:47:34 +03:00
16f3949e0b
merging with myself :p
2023-08-17 21:39:19 +02:00
a5c95c7151
Added full coqui-api model list available via new origin in Coqui UI: Coqui API (Experimental). Some model might not work, need to be tested.
2023-08-17 21:33:06 +02:00
86a486be8f
Adjust Novel instruction prompts for quiet generation
2023-08-17 17:40:38 +03:00
928c8e544f
Better random Ids for tags
2023-08-17 14:19:20 +03:00
728ffd1628
Merge pull request #957 from SillyTavern/kayrabias
...
Kayrabias
2023-08-17 13:14:05 +03:00
a9a929afaf
Fix HTML template oddity
2023-08-17 13:10:44 +03:00
0b412033ba
Streamline and simply Novel logit bias. Make it a part of preset
2023-08-17 13:05:22 +03:00
50005b5617
Reformat rvc/tts plugins
2023-08-17 12:05:17 +03:00
c4700b38fe
Merge pull request #954 from Tony-sama/staging
...
Coqui TTS add remove character voice map button + RVC upload features
2023-08-17 12:03:03 +03:00
928d897946
Fix instruct mode stop strings for whitespace sequences
2023-08-17 11:47:15 +03:00
722801bb50
Add logit biases for NovelAI
2023-08-16 22:14:04 -07:00
58123f5c29
Added RVC upload functionality, allow to send rvc model archive (zip/7z format) to extras, automatically installed properly on server side.
2023-08-17 04:47:41 +02:00
a07637aa43
Coqui TTS UI, added a button to remove choosen character from the voice map. Fixed the request spam to extras when there is no local coqui model or no rvc model in extras folder.
2023-08-17 01:16:57 +02:00
6dce566cb9
Coqui TTS UI, corrected hiding model div when origin select is none
2023-08-16 21:31:51 +02:00
3ab9b44700
Merge pull request #952 from Tony-sama/staging
...
Coqui TTS extension added support for custom local model
2023-08-16 21:50:44 +03:00
e19feddb29
Decrease max length for Novel presets
2023-08-16 21:46:58 +03:00
1f2394cada
Apply preset manager for NovelAI
2023-08-16 21:43:38 +03:00
cd097df42f
Change RVC pitch extraction "" label to "None"
2023-08-16 20:13:29 +02:00
81ed4d8431
Reorderable samplers for Novel
2023-08-16 20:34:47 +03:00
db5f815632
Introduce a state lock to prompt manager render function
2023-08-16 19:28:27 +02:00
05504ebbb2
Merge branch 'SillyTavern:staging' into staging
2023-08-16 19:19:29 +02:00
735066fe10
Added Coqui TTS extension option to use local models (no language/no speakers)
2023-08-16 19:19:13 +02:00
947289dffc
Lock up configuration during prompt manager render
2023-08-16 19:01:09 +02:00
99c3a09b8e
Reduce unnecessary console spam
2023-08-16 18:54:52 +03:00
1fa281b03c
Merge pull request #946 from Tony-sama/staging
...
RVC extension new UI and per character settings
2023-08-16 17:49:24 +03:00
3682924624
Round context messages to nearest 5 for auto-adjust
2023-08-16 17:22:42 +03:00
9e49ad0a77
Merge branch 'SillyTavern:staging' into staging
2023-08-16 14:58:14 +02:00
7a204bef05
Show chat completion source on hover
2023-08-16 13:51:38 +03:00
feb523bd01
NovelAI Kayra 1.1 update
...
* Updated some presets and added Cosmic Cube
* Change defaults for NovelAI to select Clio on cold start
* Automatically change the preset to an appropriate default whenever you change the model
* Removed deprecated Top G sampler
2023-08-15 18:52:29 -07:00
a64130c71c
Remove toastr error for character without RVC model assigned, just show it in console info to allow group chat with/withou RVC per character.
2023-08-15 21:35:07 +02:00
3cb2b7a4a9
Merge branch 'SillyTavern:staging' into staging
2023-08-15 21:15:45 +02:00
84845ae911
Updated RVC UI to allow per characters settings without manual voice map text.
2023-08-15 21:15:31 +02:00
fa35e1185e
Merge branch 'release' into staging
2023-08-15 21:51:08 +03:00
8470acda41
Add English language option
2023-08-15 21:47:49 +03:00
ba8fa8bd0c
Merge branch 'release' into staging
2023-08-15 21:38:51 +03:00
eb81c4807a
Add UI locale selector
2023-08-15 21:37:56 +03:00
4a0c93f023
Show OpenRouter icon
2023-08-15 21:36:28 +03:00
869478627d
Add UI locale selector
2023-08-15 21:32:17 +03:00
07be5588f4
Implement configurable prompt order strategy for prompt manager
...
global and character
2023-08-15 19:41:34 +02:00
e09a7d1e7a
Print stack to console on chatcompletion error
2023-08-15 19:40:22 +02:00
e9e9421d8a
Don't try to render prompt manager if group is still generating
2023-08-15 16:47:16 +03:00
57b9a94c17
Add banned tokens for Novel
2023-08-15 15:51:14 +03:00
2afac5aeb1
Filter null messages in array and context line appearing unprompted
2023-08-15 15:08:42 +03:00
706be4390d
Add new Claude instant model
2023-08-15 14:57:11 +03:00
f0d208f929
Add speed-speech due to fast processing time
2023-08-15 19:37:15 +09:00
36c6f57d96
Added stop to lip sync when user stops TTS early
2023-08-15 19:25:42 +09:00
e1ea448236
Merge pull request #939 from SillyTavern/hotfix-prompt-manager
2023-08-15 09:51:14 +03:00
b0cce17638
Minor string changes to match NAI UX
2023-08-14 19:43:05 -07:00
41ec7e5600
Many NovelAI fixes from dev guidance
...
* Remove AI Module "Autoselect" and make the auto-instruct work for all modules. This is how NAI is supposed to work.
* Log the response from the API.
* Move the AI Module setting up to the top of the settings window since it isn't part of the preset.
* Refactor phrase_rep_pen to use the actual API strings.
* Clamp the maximum token length to 150 before we call the API.
* Clamp the minimum token length in the UX from 1 to 150.
* Fix bug where the preamble was not initialized on cold start.
* Get rid of extra newline before dinkus.
* Make always_force_name2 default true.
2023-08-14 19:35:21 -07:00
9fb38af838
Fix defaults getting applied when switching back to Default preset
2023-08-15 01:28:07 +02:00
2bb8fa5ea1
Update quick edit textareas when prompt manager re-renders
2023-08-15 01:20:42 +02:00
8f830a0928
Persist settings before applying preset
2023-08-14 23:36:06 +02:00
61b645f5ff
Fix responsive popup not taking all available width on desktop
2023-08-14 23:19:39 +02:00
9f49a2612c
Return option to close chat to menu
2023-08-15 00:10:48 +03:00
a92333875c
Try to minimize side effects of generate dry running
2023-08-15 00:01:55 +03:00
238b667262
Merge branch 'staging' of https://github.com/SillyLossy/TavernAI into staging
2023-08-14 22:19:17 +03:00
3850e6b50a
Replace object stores for Chroma and token cache
2023-08-14 22:19:14 +03:00
151a13eefa
Fix formatting
2023-08-14 20:35:45 +02:00
1e71c631ae
Always show prompt manager edit popup on top
2023-08-14 20:29:09 +02:00
dbf38d359d
Debounce quick edit saving
2023-08-14 20:29:07 +02:00
033dc45ab9
Move quick edit ui identifiers into configuration
2023-08-14 20:29:01 +02:00
6caa8a4765
Add nsfw prompt to quick edit drawer
2023-08-14 20:28:57 +02:00
9543f54f8c
Fix user jb/main not replacing original placeholder in character main/jb
2023-08-14 20:28:54 +02:00
edcce96a6e
IndexedDB open sometime hangs forever
2023-08-14 19:40:31 +03:00
8c759e87cf
Better UI control for phrase rep pen
2023-08-14 19:00:36 +03:00
c172801b5e
Fix ooba's frog
2023-08-14 18:03:29 +03:00
15db16ee8d
Fuzzy search for groups
2023-08-14 17:35:48 +03:00
32745609a2
Merge pull request #936 from Tony-sama/staging
...
Refactoring of Coqui TTS extension
2023-08-14 14:00:45 +03:00
ce03921df4
Move CFG block for ooba higher
2023-08-14 13:10:44 +03:00
32eb877fdb
Add native presetable CFG scale for ooba
2023-08-14 13:06:20 +03:00
a0bfc67632
Add hint for CFG scale
2023-08-14 12:50:00 +03:00
7107813b6e
Move prompt manager down to ease mobile scrolling
2023-08-14 12:45:20 +03:00
d395d3d8e6
Add peek proxy password btn
2023-08-14 12:01:06 +03:00
6da3453efc
Fix example dialogue syntax
2023-08-14 11:52:58 +03:00
bd9f49d975
Don't launch generation intercepts on dry runs
2023-08-14 11:43:31 +03:00
e4c656a0b9
Kinda better svgs?
2023-08-14 00:09:49 -04:00
f97cc5694d
refresh character even when module is not detected
2023-08-14 05:14:21 +02:00
213099dd6d
removed debug messages
2023-08-14 05:05:10 +02:00
5816d2d6a6
Correct typo in tts index file that cause tts provider settings to be destroyed.
2023-08-14 05:03:42 +02:00
40d4123f08
Correct typo in tts index file that cause tts provider settings to be destroyed.
2023-08-14 04:59:11 +02:00
db51b44780
Rename correctly
2023-08-13 22:58:10 -04:00
c86512cf6b
Fix swiping
2023-08-13 22:57:49 -04:00
e97863a0da
Merge branch 'staging' of https://github.com/Tony-sama/SillyTavern into staging
2023-08-14 04:03:34 +02:00
c414606c81
Finished new version of coqui TTS UI.
2023-08-14 04:03:28 +02:00
fdc9349f68
Merge branch 'SillyTavern:staging' into staging
2023-08-14 04:01:32 +02:00
f528791c4f
Merge branch 'staging' of https://github.com/city-unit/SillyTavern into feature/ui
2023-08-13 18:46:37 -04:00
ad04344392
Api/Model icon option
2023-08-13 18:43:16 -04:00
3f87874afe
Merge branch 'staging' of https://github.com/Cohee1207/SillyTavern into staging
2023-08-14 05:33:48 +09:00
2935d98788
improved Escape hotkey logic
2023-08-14 05:33:46 +09:00
d14d669838
Merge pull request #932 from StefanDanielSchwarz/fixed-presets
2023-08-13 23:25:02 +03:00
21d8ce048c
Debounce more prompt manager renderings
2023-08-13 22:48:51 +03:00
e6bf870c55
Persist token cache to indexedDb
2023-08-13 22:45:27 +03:00
ef9d40428e
Update Roleplay instruct mode preset: Include Names
...
After extensive testing, I've enabled "Include Names" to fix some models talking/acting as User.
2023-08-13 20:18:22 +02:00
33a32d0380
Set TextGen temperature slider min value to 0.0
...
When updating the TextGen preset, I noticed the slider not going to the true minimum, so I fixed that as well. Same as [#825 ](https://github.com/SillyTavern/SillyTavern/issues/825 ) for KoboldAI.
2023-08-13 20:11:02 +02:00
0a4f4a6c24
Update TextGen Deterministic.settings
...
This wasn't actually deterministic because despite `do_sample: false`, temperature was still taking effect, causing non-deterministic output. I fixed this and also adjusted repetition penalty to be like KoboldAI's Deterministic preset.
2023-08-13 19:59:37 +02:00
73f9117bea
Update KoboldAI Deterministic.settings
...
After extensive testing, I've adjusted repetition penalty slightly to be the same as simple-proxy-for-tavern's default preset and ooba's LLaMA-Precise settings preset. This fixed some models talking/acting as User.
2023-08-13 19:25:56 +02:00
23a4705d36
Use common save settings function on WI budget cap
2023-08-13 19:50:12 +03:00
3688bd0b04
#930 Fix WI activation via slash command
2023-08-13 19:20:00 +03:00
43089ee06f
Don't render manager on non-chat completion APIs
2023-08-13 18:47:07 +03:00
980c65b6cf
Add toastr warning
2023-08-13 18:44:32 +03:00
c0bc425c8b
Debounce manager changes
2023-08-13 18:43:17 +03:00
9b8d069e3b
Apply CFG extension for Novel
2023-08-13 15:38:07 +03:00
bd3933bcbf
Merge pull request #905 from bdashore3/cfg-window
...
Add CFG side panel
2023-08-13 14:55:03 +03:00
9615495878
Merge pull request #768 from SillyTavern/prompt-manager-cherrypicked
...
Feature: Prompt manager (cherrypicked onto stage)
2023-08-13 14:44:28 +03:00
478bb9c59b
Enable nsfw prompt by default
2023-08-13 14:41:29 +03:00
43e91b150a
CFG: Add groupchat support
...
Groupchats have their CFG defined a little differently. Chat and
global CFGs are publicly accessible within the window. As a compromise,
individual character CFGs can be injected as part of each character's
prompt. However, these CFG settings need to be adjusted in the character's
individual chats which will carry over to the group.
In addition, make this character logic gated under a checkbox to
always prefer chat unless explicitly specified. Negative cascading
is still open, so individual character negatives can be included
at any time.
Signed-off-by: kingbri <bdashore3@proton.me>
2023-08-12 22:52:34 -04:00
e81f67504f
Merge branch 'SillyTavern:staging' into staging
2023-08-13 04:16:00 +02:00
e60f91ce42
Finished unefficient full coqui pipeline, UI generated through request to extras. Need to be pruned
2023-08-13 02:18:46 +02:00
4a21ee0337
CFG: Add ability to combine negative prompts
...
This allows for flexibility with global, character, and chat negative
prompts. Combining prompts is very useful for users who want to maintain
a set of global negatives and then add extra layers on top of that. The
ordering is chat -> character -> global tags due to the specificity of
each.
The guidance scale follows the cascade of chat -> character -> global
due to being one number that is set when CFG is fired. If the guidance
scale is 1, nothing happens.
Signed-off-by: kingbri <bdashore3@proton.me>
2023-08-12 16:25:35 -04:00
dd21091807
Selectable NovelAI module
2023-08-12 21:26:51 +03:00
0f897158b9
Create a deep clone of the preset before applying
...
Prevents non-primitives from causing overlapping references
2023-08-12 19:40:01 +02:00
f5aa7f7970
Remove obsolete prompt manager settings object
2023-08-12 18:52:52 +02:00
f434a96ad6
fix double import on char card drop
2023-08-13 01:47:33 +09:00
b0860b18b8
Show tokens when inspecting a prompt
2023-08-12 18:40:39 +02:00
03b8bb64fa
Prevent new example chat message being added for no reason
2023-08-12 18:40:39 +02:00
a9fa42524d
Fix prompts object reference being shared
...
By deepcloning a copy
2023-08-12 18:40:39 +02:00
f2c4645207
Use chatcompletion log function for token budget
2023-08-12 18:40:39 +02:00
19e5502555
Remove obsolete assignment
2023-08-12 18:40:39 +02:00
b1254fa2ab
Copy migrated preset into loaded presets
...
Since this only is populated once during page load
2023-08-12 18:40:39 +02:00
7fb896ce2a
Prevent preset migration from running multiple times.
2023-08-12 18:40:39 +02:00
a999d6e171
Fix preset migration
2023-08-12 18:40:39 +02:00
5a92f72703
Add flag to preset saving, allowing to prevent triggering the UI
2023-08-12 18:40:39 +02:00
city-unit
Cohee
Stefan Daniel Schwarz
RossAscends
deffcolony
Tony Ribeiro
Mike Weldon
maver
Tony Ribeiro
joe
SDS
kingbri