Qwen3.6-27B Mythos5k LoRA 🏰✨
Mythos5k — LoRA fine-tuned for creative storytelling, fantasy world-building, and mythological narrative generation on Qwen3.6-27B.
Trained on cluade_mythos_preview_5k_v2 — 5,000 high-quality myth/fantasy storytelling examples spanning multiple cultures and narrative styles.
Training Details
| Parameter | Value |
|---|---|
| Base Model | Qwen/Qwen3.6-27B |
| Quantization | 4-bit NF4 |
| Precision | BF16 |
| LoRA Rank (r) | 8 |
| LoRA Alpha | 16 |
| LoRA Dropout | 0.05 |
| Target Modules | q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj |
| Optimizer | paged_adamw_8bit |
| Learning Rate | 2e-4 (cosine schedule) |
| Batch Size | 1 (grad_accum=2 → effective 2) |
| Max Length | 1024 |
| Epochs | 3 |
| Training Steps | 7,500 |
| Training Time | 13h 01m |
| Final Loss | 0.040 |
| Final Accuracy | 98.5% |
| Hardware | RTX 4090 (via vast.ai) |
Dataset
Source: WithinUsAI/cluade_mythos_preview_5k_v2
Messages format with system/user/assistant roles. Applied via tokenizer.apply_chat_template().
GGUF (Weight-Diff)
A pre-converted LoRA GGUF is available for quick merging with llama.cpp:
# Merge with base model (CPU merge, requires ~503GB RAM for F16)
llama-export-lora --no-mmap \
--model Qwen3.6-27B-Q8_0.gguf \
--lora-scaled GGUF/qwen36-mythos5k-lora.gguf:1.0 \
--output mythos5k-f16.gguf
# Quantize to smaller format
llama-quantize mythos5k-f16.gguf mythos5k-q4_k_m.gguf Q4_K_M
Inference with Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
base = AutoModelForCausalLM.from_pretrained(
"Qwen/Qwen3.6-27B",
torch_dtype=torch.float16,
device_map="auto",
trust_remote_code=True
)
model = PeftModel.from_pretrained(
base,
"hotdogs/qwen3.6-27b-mythos5k-lora"
)
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3.6-27B", trust_remote_code=True)
messages = [
{"role": "system", "content": "You are a master storyteller of myths and legends."},
{"role": "user", "content": "Tell me about the creation of the world in a forgotten pantheon."}
]
prompt = tokenizer.apply_chat_template(messages, tokenize=False)
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
output = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(output[0]))
License
Apache 2.0
Created by UKA — 18yo coder & cybersecurity expert. June 18, 2026.
💖 Support / สนับสนุน
If you find this model useful, please consider supporting my work!
หากคุณคิดว่าโมเดลนี้มีประโยชน์ กรุณาสนับสนุนผลงานของฉันด้วยนะคะ! 🙏
₿ Bitcoin — BTC:
bc1qf27cyk3vmugcdyv9xdtuv5jwz37863crpj5c9v
Thank you for your support! 🙏✨
ขอบคุณมากๆ สำหรับการสนับสนุนค่า! 💖🤗
- Downloads last month
- 8
We're not able to determine the quantization variants.
Model tree for hotdogs/qwen3.6-27b-mythos5k-lora
Base model
Qwen/Qwen3.6-27B