# Kitchoon-V4-26B-A4B [*Kitchoon-V3*](https://huggingface.co/SubMaroon/Kitchoon-V3-26B-A4B) with two LoRAs baked in: [**Gemory**](https://huggingface.co/UltimateIntent/Gemory-26B-A4B-LORA) for livelier characters and my own [**Anthropos**](https://huggingface.co/SubMaroon/Anthropos-LoRA-v2-EXP) for less repetitive swipes. The Kitchoon series is designed as an RP- and **ERP**-oriented model, focusing on character development and featuring a neutral writing style.

Warning!

Thinking works, but it doesn't stop automatically.
If you use SillyTavern be sure to set Thinking Effort to any other setting (depending on Max Response Length) rather than Auto.

If you use any other engine, look for Thinking Budget.

---

Thanks to FrenzyBiscuit for helping to identify the cause of the issue with Thinking. It will be fixed in V5-Final.

## What was changed | | Kitchoon-V3 | Kitchoon-V4 | | --- | --- | --- | | Attention (`q/k/v/o_proj`) | V3 | + Gemory, all layers, at **0.5**
+ Anthropos l07, layers **13–23**, at **0.7** | | Dense MLP (`gate/up/down_proj`) | V3 | + Gemory, all layers, at **0.5**
+ Anthropos l07, layers **13–23**, at **0.7** | | MoE experts | V3 | + Gemory at **0.5** | | `lm_head`, router, embeddings, vision tower | V3 | **identical** | ---

☕ Support

If this model was useful to you, you can support my work
It helps me pay for GPU time and keep releasing new models and datasets.

Gram (TON network): UQB6cmEHWIcuiwsjgO_C0Uz4fyX3Puj09BAB3hEVVpDDcQ7K
Boosty donation: Link

--- # Details - **Gemory** ([UltimateIntent/Gemory-26B-A4B-LORA](https://huggingface.co/UltimateIntent/Gemory-26B-A4B-LORA)) — a LoRA on attention, dense MLP and MoE experts. In testing it made characters react more individually and talk to each other more. - **Anthropos l07** ([SubMaroon-exp/Anthropos-LoRA-v2-EXP](https://huggingface.co/SubMaroon-exp/Anthropos-LoRA-v2-EXP)) — my decision-point LoRA. It teaches the model to open replies and paragraphs more like the base model does, and leaves the rest of the text to the IT model. - Baked with stochastic bf16 rounding, so the baked strength matches the tested one. Verified in llama.cpp: V4 matches V3 + the merged LoRA.

Merge settings

| LoRA | Strength | Parts | Rank | |---|---|---|---| | Gemory | 0.5 | attention + dense MLP, all layers | 16 | | Gemory | 0.5 × 0.25 (correction) | MoE experts, all layers | 4 | | Anthropos l07 | 0.7 | attention + dense MLP, layers 13–23 | 16 |

## Using it > [!Note] > Recommended settings: > >| Parameter | Value | >|---|---| >| **Temperature** | `0.85` | >| **Reasoning budget** | normal or short | > > Works with and without thinking. Without thinking is the cleanest. > With thinking, prefer a normal or short reasoning budget: very long (auto) reasoning can occasionally produce odd words. ## What to expect In testing, compared to Kitchoon-V3: - swipes open differently much more often — fewer replies starting with the same words; - a different character sometimes reacts first, not always the same one; - characters react more individually: teasing, pushing back, talking to each other, not only to you; - with thinking, the reply follows the plan from the reasoning but is rewritten, not copied from the draft. Known quirks: - now and then an odd word: "own" or "quite" in the wrong place, or a doubled word ("her her"). Mostly in very long reasoning; - swipes vary in **how** the scene is delivered more than in **what happens** — the main story beat still comes from the scene; - *smirk* and the em dash are still around. ## Feedback This is an experiment, and I mostly test it on my own cards, so my coverage is narrow. If you try it, I would really like to know: - How it holds up in long sessions — scene logic, who is where, who is talking to whom. I haven't tested this enough yet. - How thinking behaves for you, and at which reasoning budget. - Whether the odd words bother you in actual play. - Your sampler settings, if you got clearly better or worse results than with the defaults. Open a thread in the Community tab, I read all of them. ## Credits Thanks to [UltimateIntent](https://huggingface.co/UltimateIntent) for Gemory, to [Tetratype](https://huggingface.co/Tetratype) for the antislop base under Kitchoon-V3, to `coder3101` for the heretic version it is built on, and to Sam Paech for the auto-antislop pipeline. Thanks to the entire `26B-Suite` team for their intellectual support. Special thanks to [Naphula](https://huggingface.co/Naphula) and [redaihf](https://huggingface.co/redaihf). You guys are awesome! **And special thanks, of course, to [Indexnusrefather](https://huggingface.co/Indexnusrefather) for his excellent dataset for original Kitchoon** **HHS1E**, **Nikita Eshkere** and all **Kitch00n ON chat** - Sending you my regards!

Full Model Card (README)