kokansei
75a1ef4304
Add DRY Samplers to ST Staging ( #2211 )
...
* Add files via upload
* Add files via upload
* Delete public/index.html
* Add files via upload
* Delete public/scripts/textgen-settings.js
* Add files via upload
* Delete public/scripts/power-user.js
* Add files via upload
* Delete public/scripts/power-user.js
* Add files via upload
* Update power-user.js
* Update index.html
* Fix control attribution
* Fix app loading
* Put sequence breakers under DRY block
* DRY for DRY
* Update public/index.html
Co-authored-by: Philipp Emanuel Weidmann <pew@worldwidemann.com >
* Merge fix
* Add llamacpp control. Add default value for sequence breakers
* Forgot reset
---------
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com >
Co-authored-by: Philipp Emanuel Weidmann <pew@worldwidemann.com >
2024-05-22 20:46:52 +03:00
kingbri
74b6ed97c2
Textgen: Add repetition decay for TabbyAPI
...
Repetition decay softens the drop off for repetition penalty. It's
best paired with rep pen range.
Signed-off-by: kingbri <bdashore3@proton.me >
2024-05-22 00:09:10 -04:00
kingbri
99d143263d
Textgen: Add skew sampling
...
Adds the option from skew sampling from exllamaV2
Signed-off-by: kingbri <bdashore3@proton.me >
2024-05-21 23:48:33 -04:00
kingbri
a12df762a0
Textgen: Add speculative_ngram for TabbyAPI
...
Speculative ngram allows for a different method of speculative
decoding. Using a draft model is still preferred.
Signed-off-by: kingbri <bdashore3@proton.me >
2024-05-21 23:37:36 -04:00
Cohee
ee913be46b
Merge pull request #2266 from sasha0552/vllm-fixes
...
vLLM fixes
2024-05-19 14:23:07 +03:00
RossAscends
c7232ae23c
WIP textgen API custom sampler display
2024-05-19 15:06:29 +09:00
sasha0552
db5e2d95c2
vLLM fixes
...
* Enable seed field for vLLM
* Enable beam search for vLLM
* Set the default length penalty to 1
(There is validation error from vLLM when beam search is disabled and the value is not equal to 1)
2024-05-19 04:34:11 +00:00
Cohee
4227968dfa
Allow using JSON schema with llamacpp server
2024-05-18 18:50:48 +03:00
Cohee
c7d75b7789
llamacpp broke
2024-05-12 21:41:07 +03:00
Cohee
27ccc6b090
Minor stylistic changes
2024-05-11 11:38:22 +03:00
kingbri
62faddac8d
Textgen: Add banned_strings
...
TabbyAPI supports the ability to ban the presence of strings during
a generation. Add this support in SillyTavern by handling lines
enclosed in quotes as a special case.
Signed-off-by: kingbri <bdashore3@proton.me >
2024-05-11 00:58:29 -04:00
Cohee
c73bfbd7b0
Safari bruh moment
2024-05-06 21:21:03 +03:00
Cohee
7063fce2af
Selectable openrouter providers
2024-05-06 19:26:20 +03:00
Cohee
05db2552b3
Fix Top K disabled state for Infermatic.
...
Also an icon.
2024-05-04 02:37:05 +03:00
Cohee
7bfd666321
Add llama 3 tokenizer
2024-05-03 23:59:39 +03:00
Cohee
7b87f44518
Clean-up API-specific settings
2024-05-03 20:02:13 +03:00
sasha0552
2bd239fe81
Initial vLLM support
2024-05-02 22:40:40 +00:00
Cohee
022c180b62
Lint and clean-up
2024-04-15 00:39:15 +03:00
Cohee
3e60919289
Specify LLM prompt in case JSON schema is not supported
2024-04-14 17:13:54 +03:00
kingbri
b8b49f0012
TextgenSettings: Fix JSON schema fallback
...
Did not fall back if the provided string was empty, resulting in
errors
Signed-off-by: kingbri <bdashore3@proton.me >
2024-04-09 22:15:00 -04:00
Cohee
51b3b8bfaa
Add smooth streaming
2024-04-02 14:56:15 +03:00
Cohee
04edf32ef0
Do not send dynatemp to backends if disabled
2024-04-02 11:29:49 +03:00
Cohee
0b76e1d350
Fix schema not loading from presets. Fix ESLint warnings
2024-04-02 11:23:29 +03:00
kingbri
5210db5679
Format
...
Signed-off-by: kingbri <bdashore3@proton.me >
2024-04-02 01:01:59 -04:00
kingbri
4f0322351e
Sampling: Add ability to send JSON schemas
...
TabbyAPI supports the ability to send JSON schemas with prompts in
addition to EBNF strings supported by outlines. Add an extra box
for TabbyAPI only.
Signed-off-by: kingbri <bdashore3@proton.me >
2024-04-02 00:59:21 -04:00
Cohee
a3ec0938c5
KoboldCpp grammar fix
2024-03-29 17:28:28 +02:00
Alexander Abushady
9bd3a526aa
Fix for unique swipes
...
Fix for unique swipes in Aphrodite
2024-03-26 23:57:24 -04:00
50h100a
8b092adc14
Use mode enum to toggle dynatemp behavior.
2024-03-25 12:25:03 -04:00
Cohee
df805d692b
Fix some code
2024-03-24 21:42:27 +02:00
50h100a
6f7e7b85ab
For Mancer:
...
- Allow logprobs (works)
- Allow multiswipe (not yet)
- Adjust visible samplers
Fix: 0 logprob is 100% chance, handle accordingly.
2024-03-24 14:45:37 -04:00
Cohee
c8f84bd413
Textgen setting refactors
2024-03-19 01:38:55 +02:00
Cohee
dc74f546d5
Merge pull request #1875 from kalomaze/cubic-curve
...
smoothing_curve UI support
2024-03-17 02:20:51 +02:00
Cohee
a0279b636b
Remove dead code
2024-03-08 08:41:54 +02:00
Cohee
2cdfda9d69
Actually use getCurrentDreamGenModelTokenizer
2024-03-08 08:40:03 +02:00
DreamGenX
bc8d41b530
Implement suggestions
2024-03-07 17:28:38 +01:00
DreamGenX
5c410986a4
Add support for DreamGen API.
...
API docs: https://dreamgen.com/docs/models/opus/v1
API keys: https://dreamgen.com/account/api-keys
I decided to base this on the text-completion API since it's more
flexible with SillyTavern's prompt formating capabilities.
This also means that custom context and instruct settings are required.
Will add documentation in a followup PR.
2024-03-07 12:25:48 +01:00
Cohee
95c49029f7
Add aphrodite model selector
2024-03-01 23:02:43 +02:00
kalomaze
45776de1d5
Smoothing curve support for ooba
2024-03-01 00:06:34 -06:00
gabriel dhimoila
76669ff8bb
add max_tokens_second
2024-02-29 00:55:25 +01:00
Cohee
f962ad5c02
Add OpenRouter as a text completion source
2024-02-25 22:47:07 +02:00
Cohee
fc289126fa
Add event type for text completion generation request settings ready
2024-02-24 21:45:33 +02:00
Cohee
d140b8d5be
Parse non-streaming tabby logprobs
2024-02-24 20:10:53 +02:00
Cohee
8848818d67
Fix dynatemp neutralization
2024-02-24 15:32:12 +02:00
Cohee
299bd9d563
Merge branch 'staging' into llamacpp-sampler-order
2024-02-24 15:10:58 +02:00
Cohee
13aebc623a
Merge pull request #1854 from deciare/llamacpp-probs
...
Request and display token probabilities from llama.cpp backend
2024-02-24 15:06:28 +02:00
Cohee
9287ff18de
Fix for non-streaming
2024-02-24 14:50:06 +02:00
Deciare
9eba076ae4
Sampler order for llama.cpp server backend
2024-02-23 23:01:04 -05:00
Deciare
936fbac6c5
Merge remote-tracking branch 'origin/staging' into llamacpp-probs
2024-02-23 17:45:54 -05:00
Deciare
344b9eedbc
Request token probabilities from llama.cpp backend
...
llama.cpp server token probabilities are given as values ranging from
0 to 1 instead of as logarithms.
2024-02-23 14:01:46 -05:00
NWilson
f569424f3e
Merge branch 'staging' into InfermaticAI
2024-02-22 08:32:10 -06:00
