Subscribe to download
AI Max ! -v0.91-

AI ! MAX !!! I’m back!!!

Aiming to be the most all-in-one, successful, useful, craziest, and well-known AI mod in the whole community.

Ofc rn it’s still kinda scuffed lol
--------------------------------------------------------------

So let me introduce this year’s biggest joke 😂

AI-Max Mod — we’re determined to be your one-stop AI support.

Including but not limited to:

- Use multiple models at the same time
- Richer model library than the official one
- Let your pet see photos! (My proud work!... maybe?)
- Supports English (haha... well)
- Multiple scenarios for active interaction (asking for food, drinks, begging to play...)
- Super ultra romantic Starry Memory Book! (All chat history will be displayed in it)
- Real streaming output, plus multiple output processing styles to choose from
- Kaomoji processing, bracket processing — if u don’t like it, u can tick special processing to show full text at once or remove it completely (this part isn’t done yet... but showing full text is ez pz)
- More customizable message modes! U can choose to play or not play the thinking animation. I promise — open the model settings, check “disable thinking animation,” and the effect might surprise u 😉
- No awkward forced “standing at attention” like the official mod (no more watching your pet snap to attention at midnight just to say one thing and go back to sleep)
- Free custom scenario prompts
- More detailed status transfer (will simplify later while keeping the effect)
- A little secret feature just for me, hehe
- Supports templated and custom system messages
- Worried your pet will “lose memory”? No worries! If you’ve used the official ChatGPT mod, you can one-click transfer memory under “Others”! (proud.png)
- Uhh ran out of stuff to flex… This version’s cover art looks good?

--------------------------------------What the official version will have------------------------------------

Ofc, version 0.91 means it’s not complete yet. Special processing only works well for the first three options (bracket processing, kaomoji processing, BetterTalk output speed). The last two (complete removal of brackets/kaomojis) are… acceptable. It runs fine though.

So, lemme lay out the roadmap:

Ugh... um... ahhhhhhh....
Ughhh.... wahhh.......

(Suddenly turns into a green-haired persona)

OK, I’m LorisYounger. What’s next?

First, DLC—jk lol
If all goes well, I’ll finish the next version tomorrow.
Testing? Why test? My code is flawless!

AHHHHHHHHHH!
WTF was that?? Ahhh I’m such a procrastinator 😭

Next version will probably take 2–3 weeks.

Planned features:
- Complete special processing
- Better multilingual support
- If everything’s good, collab with Chu Yiyun’s new mod
- If not, fall back to the special sleep function I already wrote
- Full support for Ollama local models + custom model addition (endpoint, API key, etc. — was supposed to be in this version but custom controls were too buggy, I rage quit)
- New scenarios (waking up, getting ready for sleep)
- What else? Use your imagination 😏
- Interrupt animation also interrupts requests
--------------------------------------Currently supported models------------------------------------
DeepSeek
deepseek-chat - DeepSeek V3
deepseek-reasoner - DeepSeek R1 [Long Reasoning]

Zhipu GLM
glm-4-long - GLM-4-Long [Long Context]
glm-4-air - GLM-4-AIR
glm-4-plus - GLM-4-PLUS
glm-4-flash - GLM-4-Flash
glm-4-flash-250414 - GLM-4-Flash-250414
glm-4 - GLM-4
glm-z1-airx - GLM-Z1-airX
glm-4-flashx - GLM-4-FLASHX

Volcengine Doubao
doubao-1.5-thinking-vision-pro-250428 - Doubao thinking 1.5
doubao-seed-1.6-250615 - Doubao seed1.6
doubao-seed-1.6-flash-250615 - Doubao seed1.6-flash
doubao-lite-32k-character-241015 - Doubao lite32k
doubao-lite-4k-character-240828 - Doubao lite4k [Roleplay]
doubao-character - Doubao Character

OpenAI GPT
gpt-4o - GPT-4o
gpt-4.5 - GPT-4.5
gpt-4 - GPT-4
gpt-4o-mini - GPT-4o-mini
gpt-4.1 - GPT-4.1
gpt-3.5-turbo - GPT-3.5
o1 - GPT-o1
o1-mini - GPT-o1mini

Moonlight Kimi
kimi-latest - KIMI latest
moonshot-v1-128k - KIMI moonshot v1 128k
moonshot-v1-8k - KIMI moonshot v1 8k
moonshot-v1-32k - KIMI moonshot v1 32k
moonshot-v1-128k-vision-preview - KIMI moonshot v1 128k vision
kimi-thinking-preview - KIMI thinking [bad]
kimi-k2-0711-preview - KIMI k2
kimi-k2-turbo-preview - KIMI k2 turbo

Baidu Wenxin
ernie-4.5-8k-preview - Ernie 4.5 8k
ernie-3.5-8k - Ernie 3.5 8k
ernie-3.5-8k-0613 - Ernie 3.5-0613 8k
ernie-3.5-128k - Ernie 3.5 128k
ernie-speed-128k - Ernie Speed 128k
ernie-speed-8k - Ernie Speed 8k
ernie-speed-pro-128k - Ernie Speed Pro 128k
ernie-lite-8k - Ernie Lite 8k
ernie-lite-pro-128k - Ernie Lite Pro 128k
ernie-tiny-8k - Ernie Tiny 8k

Baidu Qianfan
qianfan-8b - QianfanAI 8b
qianfan-70b - QianfanAI 70b

Tongyi Qianwen
qwen3-235b-a22b - QWEN 235b
qwen3-30b-a3b - Qianwen 3 30b

iFlytek Spark | Baichuan
Baichuan3-Turbo-128k - Baichuan Turbo3 128k
Baichuan4-Air - Baichuan 4Air
Baichuan4 - Baichuan 4
Baichuan2-53B - Baichuan 2 53B

Others
Ollama Model

Conch AI
MiniMax-Text-01 - MiniMax-Text-01
MiniMax-M1 - MiniMax-M1

Total: 53 models. Planned to reach 70 within the next 3 version updates.

--------------------------------------Gaulti's Ramblings------------------------------------

I’m burnt out.

I can finally fkin close Visual Studio and take a dump while checking how buggy this is in the community.

Mostly ‘cause it’s hard to test by myself.

Alright folks, as u can see, it’s Gaulti — I’m back, with this pile of stinky code that even I can’t stand.

Still amateur, still cringe, still shining in this community.

Actually, this mod was made for my own pet. Kinda funny — originally planned to release on 8.29 (Adoption Day), but spent the whole night fixing small bugs. Now it’s out. Also, this mod has another goal: to dethrone that R18 big-face mod that’s been stuck in the Workshop trending list (let’s not mention that part).

Let’s see if I can make it to trending.

Not afraid. Full send.

--------------------------------------How to Use------------------------------------

1. After enabling, don’t rush to talk to your pet. Open Mod Settings → Models & History → Click AddModel → Fill in your registered API and model info → Click test. If it works, a kaomoji will happily tell u “success!”
2. Then, under “SelectedModel” at the bottom of settings, choose which model u want to chat with, or select “Random” to let the system pick one for u.
3. Type what u want to say in the chatbox, hit send, and done!
4. If u want your pet to see a photo, drag the image into the message bar. If u see [filename] appear, it’s added! Current flaw: once dragged in, u can’t remove it. Next version will let u cancel.

--------------------------------------------------------------------------------------------------
Dev Bilibili: https://space.bilibili.com/1298900797
Dev Xiaohaihe: 78322840
Dev QQ: 874846980
Dev Email: GaultiVPET@163.com
If u run into issues, contact th