---
base_model: Qwen/Qwen3-8B-Base
library_name: transformers
tags:
- mergekit
- axolotl
- unsloth
- roleplay
- conversational
datasets:
- PygmalionAI/PIPPA
- Alfitaria/nemotron-ultra-reasoning-synthkink
- PocketDoc/Dans-Prosemaxx-Gutenberg
- FreedomIntelligence/Medical-R1-Distill-Data
- cognitivecomputations/SystemChat-2.0
- allenai/tulu-3-sft-personas-instruction-following
- kalomaze/Opus_Instruct_25k
- simplescaling/s1K-claude-3-7-sonnet
- ai2-adapt-dev/flan_v2_converted
- grimulkan/theory-of-mind
- grimulkan/physical-reasoning
- nvidia/HelpSteer3
- nbeerbower/gutenberg2-dpo
- nbeerbower/gutenberg-moderne-dpo
- nbeerbower/Purpura-DPO
- antiven0m/physical-reasoning-dpo
- allenai/tulu-3-IF-augmented-on-policy-70b
- NobodyExistsOnTheInternet/system-message-DPO
---

# Q3-8B-Kintsugi

FIXME: we need a new image

# Overview

***Q3-8B-Kintsugi*** is a roleplaying model finetuned from [Qwen3-8B-Base](https://huggingface.co/Qwen/Qwen3-8B-Base).

During testing, Designant punched well above its weight class in terms of parameters, especially for 1-on-1 roleplaying and general storywriting.

# Quantizations


EXL3:
- TODO!

MLX:
- TODO!

GGUF:
- TODO!
  
# Usage

- Format is plain-old ChatML (please note that, unlike regular Qwen 3, you do *not* need to prefill empty think tags for it not to reason -- see below).

- Settings used by testers varied, but Fizz and inflatebot used the same settings and system prompt recommended for [GLM4-32B-Neon-v2.](https://huggingface.co/allura-org/GLM4-32B-Neon-v2)

- The official instruction following version of Qwen3-8B was not used as a base. Instruction-following is trained in post-hoc, and "thinking" traces were not included. __As a result of this, "thinking" will not function as intended.__

# Training Process

1. The [base model](https://huggingface.co/Qwen/Qwen3-8B-Base) first went through a supervised finetune on a corpus of instruction following data, roleplay conversations, and human writing based on the [Ink](https://huggingface.co/collections/allura-org/ink-6772fd1442308781594bbabb)/[Bigger Body](https://huggingface.co/collections/allura-org/bigger-body-67b277af0861cec33b54745d)/[Remnant](https://huggingface.co/collections/allura-org/remnant-6817c2113bbb2aed501513d0) lineage.

2. Finally, a KTO reinforcement learning phase steered the model away from the very purple prose the initial merge had, and improved its logical+spatial reasoning and sense of overall "intelligence".

Both stages here are very similar to [Q3-30B-A3B-Designant](https://huggingface.co/allura-org/Q3-30B-A3B-Designant), which went through a very similar process with the same data.

# Credits

- Fizz - Training, Data Wrangling

- Toaster, Mango, Bot, probably others I forgot ;-; - Testing

- inflatebot - original Designant model card that this one was yoinked from

- Artus - Funding

- Alibaba - Making the original model

- Axolotl, Unsloth, Huggingface - Making the frameworks used to train this model (Axolotl was used for the SFT process, and Unsloth+TRL was used for the KTO process)

We would like to thank the Allura community on Discord, especially Curse, Heni, Artus and Mawnipulator, for their companionship and moral support. You all mean the world to us <3

---
<sup>*`There, God is not.`*</sup>