lite.koboldai.net Open in urlscan Pro
2606:4700:3033::6815:5ee2  Public Scan

Form analysis
0 forms found in the DOM

Text Content

 * Reconnect
 * Custom Endpoint
 * AI
 * New Session
 * Scenarios
 * Quick Start
 * Save / Load
 * Settings

Connected to AI Horde


Welcome to KoboldAI Lite!
You are using the models aphrodite/NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS,
aphrodite/NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS,
koboldcpp/dolphin-2.9-llama3-8b-q6_K, koboldcpp/Fimbulvetr-11B-v2,
koboldcpp/Fimbulvetr-11B-v2.Q4_K_M, koboldcpp/LLaMA2-13B-Psyfighter2 and 12
others.
There are 73 volunteer(s) running selected models with a total queue length of
81460 tokens.

Instruct Mode Selected - Enter a prompt below to begin!
Or, load a JSON File or a Character Card here.
Or, select a Quick Start Scenario here.



Context Back Redo Retry Add Img
Enter Sends
Allow Editing

Story
Chat
Select

Submit [ABORT]
KoboldAI Lite - A frontend for self hosted and third party API services
The AI is typing...

Context Back Redo Retry Add Img Edit

KoboldAI Lite - A frontend for self hosted and third party API services.
Quick Start - Select A Scenario
All Story Adventure Chat Instruct
Automatically select AI model ? This option picks a suitable AI model based on
the selected scenario. If no text model is currently selected, an appropriate
one will be automatically picked for you.
Allow NSFW Models ? If disabled, NSFW only models like Erebus will never be
selected


Ok Cancel
Save File / Load File / Export File

Caution: Storage Slots are saved to a tempoary cache and can be deleted by your
browser. To avoid losing data, use the download file button.

Back
Select your AI provider
AI Horde KoboldAI Remote API OpenAI API OpenRouter API Claude By Anthropic API
PaLM/Gemini By Google API Cohere API
The AI Horde is a service that generates text using crowdsourced GPUs run by
independent volunteer workers. Avoid sending privacy sensitive information.
Click here for more info
Your AI Horde API Key ? You need an API key to use AI Horde to generate text.
Get one at https://aihorde.net/register or use the anonymous key 0000000000.
[Manage My Workers] Need a Key?
(Register New User)
Select AI Horde Model ? These are the models currently provided by AI Horde
volunteers. [See Current Volunteers]
Select By Worker ? This option explicitly assigns worker IDs, fixed based on the
current workers available at model selection time.
You can use this to connect to a KoboldAI instance running via a remote tunnel
such as trycloudflare, localtunnel, ngrok.

Localhost IPs require host mode enabled. You can use the remote address
displayed in the remote-play.bat window or colab window, note that the model
must be loaded first.

Please input URL of the KoboldAI instance.



Show Console Logging
Entering your OpenAI API key will allow you to use KoboldAI Lite with their API.

Note that KoboldAI Lite takes no responsibility for your usage or consequences
of this feature. Your API key is used directly with the OpenAI API and is not
transmitted to us.
Only Temperature, Top-P and Repetition Penalty samplers are used.

Please input OpenAI API URL and Key.

Entering your OpenRouter API key will allow you to use KoboldAI Lite with their
API.

Note that KoboldAI Lite takes no responsibility for your usage or consequences
of this feature. Your API key is used directly with the OpenRouter API and is
not transmitted to us.
Only Temperature, Top-P and Repetition Penalty samplers are used.

Please input OpenRouter Key.


Model Choice:
gpt-3.5-turbo-instruct davinci-002 gpt-3.5-turbo gpt-3.5-turbo-16k gpt-4
gpt-4-turbo gpt-4o gpt-4-32k [Custom] openai/gpt-3.5-turbo openai/gpt-4
openai/gpt-3.5-turbo-instruct mistralai/mistral-7b-instruct
gryphe/mythomax-l2-13b huggingfaceh4/zephyr-7b-beta anthropic/claude-2 [Custom]
Fetch List Use Custom
Add Endpoint Version
Use ChatCompletions API

Main Message Role: User Assistant System
Add Prefix
Add Postfix
User Assistant System
User Assistant System
Entering your Claude API key will allow you to use KoboldAI Lite with their API.

Note that KoboldAI Lite takes no responsibility for your usage or consequences
of this feature.
Only Temperature, Top-P and Top-K samplers are used.

NOTICE: At this time, the official Claude API has CORS restrictions and must be
accessed with a CORS proxy. Your connection WILL be proxied.

Please input Claude API URL and Key.


Model Choice:
claude-v1 claude-v1-100k claude-instant-v1 claude-instant-v1-100k claude-2
claude-2.1 claude-2.0 claude-3-opus claude-3-sonnet claude-3-haiku
Add Endpoint Version
Claude Compatibility Rename Fix
Uses Gemini or PaLM Text Bison by Google.

Note that KoboldAI Lite takes no responsibility for your usage or consequences
of this feature.

gemini-pro gemini-1.5-pro-latest gemini-1.5-flash-latest text-bison-001 Please
input Gemini or PaLM API Key.




Uses Cohere's models through their own API.

Note that KoboldAI Lite takes no responsibility for your usage or consequences
of this feature.

command command-r command-r-plus Please input Cohere API Key.



Use WebSearch
Use Preamble
Ok Cancel
Really Start A New Story?
Unsaved data will be lost.


Keep AI Selected?
Keep Memory and World Info?

Ok Cancel
Advanced Load File
Select categories to import from saved file. Selected categories will be
overwritten. Unselected categories will retain original values.



Main Story Memory and Author's Note World Info Stop Sequences General Settings
Aesthetic Settings

Ok Cancel
Image Information

Loading...

Delete Image Close
Settings
 * Basic
 * Advanced

Quick Presets ?Pick from an easy selection of curated generation presets, or
configure your own.
[Default]
Temperature ?Randomness of sampling. High values can increase creativity but may
make text less sensible. Lower values will make text more predictable but can
become repetitious.

0.1
2
Max Ctx. Tokens ?Max number of context tokens submitted to the AI. Must exceed
Amount to Generate. Can be further increased by editing the textbox. Older
models stop at 2048, newer ones can do 4096 or greater.

512
2048
Auto-Adjust Limits

Amount to Gen. ?Number of tokens the AI should generate. Higher numbers will
take longer to generate.

16
512
Auto-Adjust Limits
Top p Sampling ?Used to discard unlikely text in the sampling process. Lower
values will make text more predictable but can become repetitious. Set to 1 to
deactivate it.

0
1
Repetition Penalty ?Used to penalize words that were already generated or belong
to the context (Going over 1.2 breaks 6B models).

1
3
Format ?Story Mode is best for novel style writing. Adventure Mode is best for
Interactive Fiction RPGs. Chat Mode is best for chat conversations with the AI.
Instruct mode is for giving the AI ChatGPT styled tasks.
Instruct Mode Story Mode Adventure Mode Chat Mode
UI Style Select ?Select your preferred UI style, which affects text formatting
and display. Some UIs are only available for specific modes.
Classic Messenger Aesthetic ⚙️
Chat PrePrompt ?Modifies the context, injecting tokens to improve chat quality
for new chats.
Inject Timestamps ?Injects timestamps into context, allowing the AI to have a
sense of time.
Adventure PrePrompt ?Modifies the context, injecting tokens to improve adventure
quality for new adventures.
Enable Markdown ?Allows the UI to use markdown formatting such as quotes and
code blocks.
Inject Timestamps ?Injects timestamps into context, allowing the AI to have a
sense of time.
Inject ChatNames ?Appends chat names after every instruct tag, a hybrid chat
mode.
Assistant Jailbreak ?Automatically injects a jailbreak message after every query
to make the AI more likely to obey you.
Multiline Replies ?Whether to allow multiple lines in AI responses. Disable this
if the AI starts generating rubbish.
Instruct Tag Preset ?Quickly select between common instruct tag formats.
Different models are trained with different tags.
[Custom] Alpaca Vicuna Metharme Llama 2 Chat Q & A ChatML KoboldAI Format
CommandR Llama 3 Chat
Sys. Prompt ?A system pre-prompt sent at the very start to guide the AI
behavior. Usually NOT needed.

Start Seq.?The sequence to start an instruction prompt End Seq.?The sequence to
end an instruction prompt

Your Name AI Name ?Name of the person(s) you want to chat with. Multiple
opponents can be specified, creating a group chat, separate their names using
multiple lines.

Multiline Replies ?Whether to allow multiple lines in AI responses. Disable this
if the AI starts generating rubbish.
Continue Bot Replies ?Allow incomplete AI chat replies, which can be continued
by pressing submit again. Not recommended for newbies.
Advanced Sampler Config ?These settings control alternative samplers
configurations. They are inactive by default, you usually do not need to change
them.

Top-K Top-A Typ. TFS Seed Min-P PrPen. DyTmp. OFF

Mirostat (If supported) ?Replaces your samplers with mirostat, an alternative
sampling method. May not be available depending on backend, not supported on
Horde.

Mode Tau Eta Off 1 2

Mirostat Not Supported
User Mods ?Allows you to load third-party user created mods (caution).
Apply User Mod

RpRng. RpSlp. Smp.Order ?The order by which all 7 samplers are applied,
separated by commas. 0=top_k, 1=top_a, 2=top_p, 3=tfs, 4=typ, 5=temp, 6=rep_pen

Additional Configs ?Grammar Sampling (KCPP) - Allows you to constrain output to
fit specific structures. Resets grammar state every generation unless Retain is
checked.
Set Grammar
Retain
Generate Images ?Use the AI Horde or a local A1111 instance to insert AI
generated images into your story.
[Disabled] AI Horde Local A1111 OpenAI DALL-E
Set Horde Key
Allow NSFW
[None] ⚙️
Save In A1111

Set URL Set Key

Autogenerate
Save Images
Setup
🎨

TTS ?Enable Text-To-Speech to have your story automatically read to you.
Test

Set URL female_calmfemalemale Language

Narrate Both Sides
Narrate Only Dialog
Beep on Done
Notify on Done
Reset ALL Settings
Token Streaming ?Use token streaming for partial responses. SSE is smoother but
less well-supported. Poll is chunkier but more reliable. Not available on Horde.
Off Poll SSE
Idle Responses 
Off 1x 2x 3x 5x 8x 10x 15s 30s 60s 2m 5m 10m Auto
Trim Sentences ?Trims incomplete sentences in AI output.
Trim Whitespace ?Removes trailing whitespace in AI output.
Compress Newlines ?Compresses multiple newlines into one newline in AI output.
EOS Token Ban ?Allow the End-Of-Stream (EOS) token and potentially other
restricted special tokens to be generated.
Auto Unban Ban Bypass
Placeholder Tags ?If enabled, uses universal {{user}} and {{[INPUT]}}
placeholders that get swapped on submit. If disabled, uses plaintext chat or
instruct tags verbatim.
Render Sp.Tags ?If enabled, renders special tags like EOS and padding tokens.
Not recommended.
Run In Background ?Prevents the browser from suspending Kobold Lite by playing a
silent audio track. This setting cannot be saved.
Autosave Session ?Autosaves your current story and settings on exit, reloads
when you return
Embed Settings File ?Includes your current settings when saving or sharing your
story
Rename Save File ?Prompts to input a different filename when saving file.
Show Advanced Load ?If enabled, allows you to select additional configurations
during file load
Autoscroll Text ?Automatically scrolls the text window down when new text is
generated
Unlock Scroll Height ?Unlocks the text viewport, allowing for infinite height
without scrolling
Viewport Width ?Controls horizontal scaling of the viewport window
Adapt Clamp HDClamp Unlock
Inverted Colors ?Inverts all colors, simple light mode
Background Img
Set Clear
OK Cancel
Context Data
 * Memory
 * World Info
 * Tokens

Memory?Put the information you want the AI to always remember. It will be
inserted into the top of every request sent to the AI.
Newline After Memory

Author's Note?Similar to Memory, but inserted near the end of the text instead
of the start. A good way to control the mood/behavior of the AI.
Notes AutoGenerate Memory

Author's Note Template?A placeholder, will be inserted with the author's note
replacing the <|>. You generally don't need to change this. A/N
Strength?Controls how far back to insert the Author's Note. Notes injected
closer to the end have a stronger effect.
Weak Medium Strong Immediate
+Add

WI Insert Location ?Controls where the world info should be inserted
After Memory Before A/N
WI Search Depth ?Controls how far back in the text to search for World Info Keys
Full Context Last 1024 Last 512 Last 256

Case Sensitive Keys
Extra Stopping Sequences ?Triggers the text generator to stop generating early
if this sequence appears, in addition to default stop sequences. If you want
multiple sequences, separate them with the following delimiter: ||$||
Stop Sequences may be unavailable.
Add New
Logit Biases ?Specify a dictionary of token IDs to modify the probability of
occuring. Expand Section
Logit bias may be unavailable.
Enter OpenAI-formatted logit bias dictionary. Each key is the integer token IDs
and their values are the biases (-100.0 to 100.0). Leave blank to disable.
Input is a JSON object, reference here.

Add New
Token Filter ?Outright removal for ANY tokens containing a specific substring
from model vocab. If you want multiple sequences, separate them with the
following delimiter: ||$|| Expand Section
Token filter may be unavailable.
Outright removal for ANY tokens containing a specific substring from model
vocab. If you want multiple sequences, separate them with the following
delimiter: ||$||
Note: If you're trying to ban a specific token ID, you should use Logit Bias
instead!

Add New
Regex Replace ?Allows transforming incoming text with regex patterns, modifying
all matches. Replacements will be applied in sequence. Expand Section



Placeholder Tags ?Configure automatic substitutions for placeholders in text.
Expand Section
Stories can use placeholders like {{user}} and {{[INPUT]}} that require dynamic
substitution. If disabled, uses plaintext tags verbatim.
Enable Placeholder Tags ?If enabled, uses placeholders that get swapped on
submit. If disabled, uses plaintext verbatim.



OK Cancel
Worker List

NameModelCapabilitiesUptimeKudos

OK
My Worker List

NameDescriptionUptimeKudosMaint.Del.

OK
Share Story

Copy Close
Warning: This story is very long. It may not load in some browsers. You should
save it as a file instead.
Dynamic Temperature Wizard
Dynamic temperature is specified by a Temperature Value and a Temperature Range.
Actual temperature is allowed to be automatically adjusted dynamically between
(DynaTemp ± DynaRange).

For ease of use, a simple converter is provided here. Setting both values to the
same temperature disables DynaTemp.


Minimum Temperature:
Maximum Temperature:

--------------------------------------------------------------------------------

Temperature:
DynaTemp-Range:

--------------------------------------------------------------------------------

DynaTemp-Exponent:
Smoothing-Factor:
Ok
Image Generation Settings
Style tags to use for generating images:
(E.g. Sketch, Realistic, Anime, 3D Render, Drawing)

Negative Prompt

Number of Steps:
Cfg. Scale:
Sampler:
Euler A Euler Heun DPM2 LCM DPM++ 2M
Aspect Ratio ? Square is recommended. Changing aspect ratio will affect the
resolution used to generate. This may impact quality or memory usage. :
Square Portrait Landscape
Img2Img Strength ? Higher values lead to a more different image. :
Save Higher-Res ? This option will result in larger save files which may be
slower. Changing this setting only applies to NEW images. :
Ok
Add New Image
Generate Image (Automatic)
Generate Image (Custom Prompt)
Upload Image File
Paste from Clipboard
Customize Settings
Go To StableUI

Cancel
Paste Image From Clipboard

Cancel
Share Story Import / Export
Export Share as TextData
Export Share as Web URL
Import Share from TextData

Cancel
Chat Selectors

Ok Cancel
Aesthetic UI customization panel
Background Style
Bubble Color:
System 🖌️
You 🖌️
AI 🖌️
Rounded Bubbles:
Color Background:
Min Height:
px
Horizontally-centered text:
Margin (px):
L:
R:
T:
B:
Padding (px):
L:
R:
T:
B:
Portrait Style
Portraits:
🖼️ Your Portrait
🖼️ AI's Portrait
Portrait Style:
None Circle Rounded Rect
(Reset Image)
User Portrait:
Size:
px
A/R:
AI Portrait:
Size:
px
A/R:
Show Names (Chat Mode):
Font Style
Font Size:
px
Customize:
Per-entity:
Style Text:
Colors:
text🖌️
"speech"🖌️
*action*🖌️
You:
text🖌️
"speech"🖌️
*action*🖌️
AI:
text🖌️
"speech"🖌️
*action*🖌️
System:
text🖌️
"speech"🖌️
*action*🖌️
Code blocks:
background🖌️
foreground🖌️

(Reset All Styles)
OK Cancel

Style Preview

You are Mikago, a prestigious bot that's a supervillain.

Roleplay in first person, be prestigious, don't be a bot. This is a fantasy
world.

Code blocks should be wrapped in triple backticks, like so:




-- multiline
--- code here$




takes my hat off to greet the squad “Greetings, I am Mikago, the prestigious!”
bows to the crew
clears my throat “Now, I'm sure there are many questions, but all will be
answered in due time.” deep breath
draws my sword “Yes. You should know the code to calculate the factorial of a
number.”
The crew also draws their weapons and point them at you, not giving you any
space.
backs off “Woah, easy there..” makes some steps backwards, but then stops
“I would normally take this as an insult to my prestige, but I understand your
caution..” takes a deep breath
“Well, if it's to prove myself, here goes the python code to calculate the
factorial of a number..”

Mikago opens a live-code-portal with his magic and writes the code that was
requested.




def factorial(n):
  if n == 0:
    return 1
  else:
    return n * factorial(n-1)



looks at you, getting impatient “Are we ok now.. or do you want me to write the
code of a game next?”

sheathes my sword and approaches for a hug “Oh, Mikago, my old friend, it is
really you!”



Yes No
OK Cancel
OK