qwen3-ascii-tui-lora

qwen3-ascii-tui-lora

LoRA adapter for Qwen/Qwen3-0.6B specialized for terminal-first educational content:

• high-quality ASCII diagrams
• TUI-style process maps
• concise conceptual explanations paired with diagram blocks

Model Details

• Developed by: mr-dee
• Base model: Qwen/Qwen3-0.6B
• Fine-tuning method: QLoRA / PEFT
• Adapter repo: mr-dee/qwen3-ascii-tui-lora
• Source code: https://github.com/dylanler/ascii-tui-qwen3-lab

Intended Use

Direct use:

• Generate CLI-friendly educational diagrams for science and engineering topics.
• Produce compact text layouts suitable for terminals, docs, and chat interfaces.

Downstream use:

• Learning tools, documentation assistants, terminal tutors, and curriculum prototyping.

Out-of-scope:

• Safety-critical decisions.
• Legal/medical/financial advice.
• Fully automated scientific tutoring without human review.

Training Data

Dataset characteristics:

• 1,000 synthetic instruction-response rows
• 950 train / 50 eval split
• 40 unique topics
• Required topic coverage includes:
  • lifecycle of a volcano
  • how does gravity work
  • double-slit experiment

Each row includes:

• topic
• instruction
• response (multiline explanation + ASCII/TUI diagram)
• diagram_style, difficulty, and tags

Training Procedure

Hardware:

• 4x NVIDIA A100-SXM4-80GB (DDP via torchrun --nproc_per_node=4)

LoRA config:

• rank r=32
• alpha 64
• dropout 0.05
• target modules: q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj

Optimization:

• epochs: 3
• learning rate: 2e-4
• per-device train batch size: 4
• gradient accumulation: 4
• max sequence length: 1024
• mixed precision: bf16
• optimizer: paged AdamW 8-bit

Metrics Snapshot

From trainer_state.json and train_metrics.json:

• train points: 45
• eval points: 9
• train loss: 4.12497 -> 0.10813
• eval loss: 2.06111 -> 0.11857
• best checkpoint: checkpoint-45
• train runtime: 88.83s
• train run average loss: 0.69445

Quick Start

Transformers + PEFT

import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

BASE = "Qwen/Qwen3-0.6B"
ADAPTER = "mr-dee/qwen3-ascii-tui-lora"

tokenizer = AutoTokenizer.from_pretrained(BASE, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    BASE,
    torch_dtype=torch.bfloat16,
    device_map="auto",
    trust_remote_code=True,
)
model = PeftModel.from_pretrained(model, ADAPTER)

messages = [
    {"role": "system", "content": "You create terminal-friendly educational ASCII/TUI diagrams."},
    {"role": "user", "content": "Explain the double-slit experiment with a compact ASCII diagram."},
]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=600, temperature=0.2, top_p=0.9)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))

vLLM (OpenAI-compatible endpoint)

vllm serve Qwen/Qwen3-0.6B \
  --enable-lora \
  --lora-modules ascii=mr-dee/qwen3-ascii-tui-lora \
  --tensor-parallel-size 2 \
  --max-model-len 2048

Query example:

curl http://127.0.0.1:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "ascii",
    "messages": [
      {"role":"user","content":"Teach gravity using an ASCII flow diagram and 3 takeaways."}
    ],
    "max_tokens": 500,
    "temperature": 0.2
  }'

Important:

• keep prompt_tokens + max_tokens <= max_model_len
• if exceeded, vLLM returns HTTP 400 validation errors

Risks and Limitations

• May generate plausible but incorrect scientific claims.
• ASCII layout quality can degrade for very long outputs.
• Not guaranteed to follow strict width limits unless prompted.
• Should be reviewed by a human before educational publication.

Responsible Use

• Verify scientific correctness.
• Validate generated text/code before classroom use.
• Avoid over-reliance in high-stakes domains.

Framework Versions

• PEFT: 0.18.1
• Transformers: 4.57.6
• TRL: 0.27.2
• PyTorch: 2.9.1

Citation

@misc{qwen3_ascii_tui_lora_2026,
  title  = {qwen3-ascii-tui-lora},
  author = {mr-dee},
  year   = {2026},
  url    = {https://huggingface.co/mr-dee/qwen3-ascii-tui-lora}
}