---
base_model: nbeerbower/Xiaolong-Qwen3-0.6B
datasets:
- nbeerbower/GreatFirewall-DPO
- nbeerbower/Schule-DPO
- nbeerbower/Purpura-DPO
- nbeerbower/Arkhaios-DPO
- jondurbin/truthy-dpo-v0.1
- antiven0m/physical-reasoning-dpo
- flammenai/Date-DPO-NoAsterisks
- flammenai/Prude-Phi3-DPO
- Atsunori/HelpSteer2-DPO
- jondurbin/gutenberg-dpo-v0.1
- nbeerbower/gutenberg2-dpo
- nbeerbower/gutenberg-moderne-dpo
- GeneralReasoning/GeneralThought-430K
- nvidia/OpenMathReasoning
- nvidia/OpenCodeReasoning
library_name: transformers
license: apache-2.0
tags:
- orpo
- uncensored
- reasoning
- cot
- llama-cpp
- gguf-my-repo
---

# Triangle104/Xiaolong-Qwen3-0.6B-Q4_K_S-GGUF
This model was converted to GGUF format from [`nbeerbower/Xiaolong-Qwen3-0.6B`](https://huggingface.co/nbeerbower/Xiaolong-Qwen3-0.6B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/nbeerbower/Xiaolong-Qwen3-0.6B) for more details on the model.

---
Xiaolong is a small, uncensored, reasoning-focused model finetuned using ORPO and QLoRA on top of Qwen3-0.6B-abliterated-TIES.

Finetuning Details
-

- Method: ORPO
- Epochs: 1.3
- Learning Rate: 5e-6, cosine decay w/ 5% warmup
- Batch Size: 4 x 8 (32 effective)
- Max Grad Norm: 0.3
- LoRA Rank: 64
- Hardware: 1x NVIDIA RTX A6000

Dataset Composition
-
~9,100 samples. 3,000 used Chain of Thought reasoning.

- nbeerbower/GreatFirewall-DPO
- nbeerbower/Schule-DPO
- nbeerbower/Purpura-DPO
- nbeerbower/Arkhaios-DPO
- jondurbin/truthy-dpo-v0.1
- antiven0m/physical-reasoning-dpo
- flammenai/Date-DPO-NoAsterisks
- flammenai/Prude-Phi3-DPO
- Atsunori/HelpSteer2-DPO (1000 samples)
- jondurbin/gutenberg-dpo-v0.1
- nbeerbower/gutenberg2-dpo
- nbeerbower/gutenberg-moderne-dpo

Chain of Thought
-
- GeneralReasoning/GeneralThought-430K (1000 samples)
- nvidia/OpenMathReasoning (1000 samples)
- nvidia/OpenCodeReasoning (1000 samples)

---
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)

```bash
brew install llama.cpp

```
Invoke the llama.cpp server or the CLI.

### CLI:
```bash
llama-cli --hf-repo Triangle104/Xiaolong-Qwen3-0.6B-Q4_K_S-GGUF --hf-file xiaolong-qwen3-0.6b-q4_k_s.gguf -p "The meaning to life and the universe is"
```

### Server:
```bash
llama-server --hf-repo Triangle104/Xiaolong-Qwen3-0.6B-Q4_K_S-GGUF --hf-file xiaolong-qwen3-0.6b-q4_k_s.gguf -c 2048
```

Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```

Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```

Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Triangle104/Xiaolong-Qwen3-0.6B-Q4_K_S-GGUF --hf-file xiaolong-qwen3-0.6b-q4_k_s.gguf -p "The meaning to life and the universe is"
```
or 
```
./llama-server --hf-repo Triangle104/Xiaolong-Qwen3-0.6B-Q4_K_S-GGUF --hf-file xiaolong-qwen3-0.6b-q4_k_s.gguf -c 2048
```