Skip to content

MoonshotAI Moonshot Behavior

This page covers the toggles and options that control how IntenseRP interacts with Moonshot (kimi.com).

Model IDs

Moonshot exposes three API behavior presets: moonshot-auto, moonshot-chat, and moonshot-reasoner. These are behavior modes, not separate backend model selection.


Authentication

Kimi uses a Google sign-in popup rather than a normal in-page email/password form.

IntenseRP can now try a best-effort Auto Login flow there by pasting your saved Google email/password into the popup and pressing Enter between steps.

That said, Google is still Google. It may decide to ask for extra confirmation, show an account chooser, demand 2FA, or just leave the popup hanging there. When that happens, IntenseRP falls back to manual completion and waits for you to finish the flow.

Persistent Sessions are still strongly recommended. If your Google session stays alive, Kimi becomes much less annoying to use.

Multiple profiles

There's rate limiting as well, so you can still set up multiple Kimi accounts/profiles.

Open Settings -> Provider and Login -> Sign-In and Accounts -> Saved Accounts and add multiple accounts under Moonshot.

IntenseRP uses those values for Moonshot's Google popup Auto Login when Auto Login is enabled, and each row also gets its own browser profile/session.

You may wish to log in with different Google accounts in each profile to further reduce the risk of rate limits.

Thinking

Kimi exposes reasoning through model mode selection in the web UI.

Enable Thinking

Switches Kimi to K2.6 Thinking before sending a request. When disabled, IntenseRP switches to K2.6 Instant.

Settings -> Provider Behavior -> Moonshot -> Enable Thinking

Send Thinking

When enabled, reasoning content is included in API output, wrapped in <think> tags. When disabled, only final answer text is forwarded.

Settings -> Provider Behavior -> Moonshot -> Send Thinking

Model downgrading

As part of a paywall strategy, Kimi will downgrade to K2.6 Instant (if you're on their Free plan) when you enable Thinking mode. This is a provider-side change and not something IntenseRP can control. They say it's because of high demand, but really it probably just means they're trying to force you to pay ¯\_(ツ)_/¯.


Toggles Kimi search tooling in the web UI.

Settings -> Provider Behavior -> Moonshot -> Enable Search

Search + Thinking

Kimi can emit multi-stage reasoning when Search and Thinking are both enabled. Some clients (including SillyTavern) may not parse this perfectly.

On a personal note

Do not enable Search and Thinking at the same time. Just don't. It's a mess and IntenseRP sometimes breaks trying to handle it.


Provider guardrails (automatic)

Kimi's Memory feature can inject account-level context into chats.

To keep chats predictable, IntenseRP watches Kimi's startup settings response and auto-disables Memory if it sees it enabled.

This is done from the browser context, so it uses your active Kimi session and cookies rather than a separate HTTP client.


File Upload Mode

Instead of typing your message into Kimi's editor, IntenseRP can upload it as a text file attachment.

Settings -> Provider Behavior -> Moonshot -> Send As Text File

File Upload Timeout

Controls how long IntenseRP waits (in seconds) for the send button to become enabled after upload.

Settings -> Provider Behavior -> Moonshot -> File Upload Timeout

Default is 15 seconds.

Text File Filler

Kimi won't let you send a file with an empty textbox as it needs some text alongside it. By default IntenseRP pastes a single . (dot) as filler, but you can change this to whatever you want.

Settings -> Provider Behavior -> Moonshot -> Text File Filler

This setting only appears when Send As Text File is enabled.


Anti-Censorship

When enabled, IntenseRP suppresses refusal-like stream events (when detected) and closes the response cleanly.

Settings -> Provider Behavior -> Moonshot -> Anti-Censorship

What It Doesn't Do

This does not bypass provider filtering. It only suppresses refusal-style output from being forwarded to the client.


Reuse Matching Chat

When enabled, IntenseRP tries to click Kimi's regenerate action if:

  1. The new prompt is identical to the cached last prompt
  2. Effective behavior settings are also identical

Otherwise it starts a fresh chat.

Settings -> Provider Behavior -> Moonshot -> Reuse Matching Chat

Search Older Matching Chats

If you enable Provider Behavior -> Moonshot -> Search Older Matching Chats, IntenseRP keeps up to 7 older cached Kimi chats per account instead of only remembering the latest one.

When one of those older chats matches the current prompt/settings, IntenseRP can reopen it and use Kimi's regenerate action there.


Delete Chat After Reply

If you want Kimi's chat list to stay tidier, IntenseRP can delete the completed chat after a successful reply finishes.

Settings -> Provider Behavior -> Moonshot -> Delete Chat After Reply

Slower requests

This adds extra cleanup work after each request, so it can slow requests down quite a bit.

No chat reuse here

This does not work together with Reuse Matching Chat or Search Older Matching Chats.

See also: Chat Auto-Deletion


Per-Message Macros

You can add [[...]] macros to the latest user message to override behavior for that request only. All macros are stripped before sending.

Macro Effect
[[think]], [[r1]] Force Thinking on
[[nothink]], [[r0]] Force Thinking off
[[search]] Force Search on
[[nosearch]], [[no_search]] Force Search off
[[file]] Force Send As Text File on
[[nofile]] Force Send As Text File off

Quick Reference

Setting What It Does Default
Enable Thinking Switches Kimi mode between Instant and Thinking Off
Send Thinking Includes reasoning in response Off
Enable Search Toggles Kimi search Off
Send As Text File Uploads prompt as .txt Off
File Upload Timeout Seconds to wait for upload 15
Text File Filler Text pasted alongside the uploaded file .
Anti-Censorship Suppresses refusal-like outputs Off
Reuse Matching Chat Regenerates on duplicate prompts Off
Delete Chat After Reply Deletes the completed Kimi chat after a successful reply Off

Back to Providers

Providers Overview