Instructions to use issdandavis/polly-chat-qwen-0.5b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use issdandavis/polly-chat-qwen-0.5b with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")
model = PeftModel.from_pretrained(base_model, "issdandavis/polly-chat-qwen-0.5b")

Transformers

How to use issdandavis/polly-chat-qwen-0.5b with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="issdandavis/polly-chat-qwen-0.5b")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("issdandavis/polly-chat-qwen-0.5b", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use issdandavis/polly-chat-qwen-0.5b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "issdandavis/polly-chat-qwen-0.5b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "issdandavis/polly-chat-qwen-0.5b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/issdandavis/polly-chat-qwen-0.5b

SGLang

How to use issdandavis/polly-chat-qwen-0.5b with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "issdandavis/polly-chat-qwen-0.5b" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "issdandavis/polly-chat-qwen-0.5b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "issdandavis/polly-chat-qwen-0.5b" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "issdandavis/polly-chat-qwen-0.5b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use issdandavis/polly-chat-qwen-0.5b with Docker Model Runner:
```
docker model run hf.co/issdandavis/polly-chat-qwen-0.5b
```

polly-chat-qwen-0.5b / README.md

issdandavis

Update model card: mark as pre-pump baseline, point to v2 training

1912e05 verified 6 months ago

1.08 kB

language:
  - en
license: mit
tags:
  - polly
  - scbe-aethermoore
  - sacred-tongues
  - sft
  - qwen2.5
  - qlora
base_model: Qwen/Qwen2.5-0.5B-Instruct

Polly Chat (Qwen 0.5B) — Pre-Pump Baseline

Early baseline Polly chatbot fine-tuned on SCBE-AETHERMOORE lore data.

Status: Superseded. A new version trained on 233K multi-view pairs with pump orientation (L0/L1/L2/L3 tasks) is in progress.

What This Model Is

QLoRA fine-tune of Qwen2.5-0.5B-Instruct on early Spiralverse lore data. This was the first Polly training run before the pump architecture existed.

What Comes Next

Polly v2 (baseline): Qwen2.5-3B on 116K L3-only pairs
Polly v2 (stack-lite): Qwen2.5-3B on 233K multi-view pairs (L0+L1+L2+L3)
Pump orientation: inference-time tongue profiles + null patterns + governance

Training Data

Dataset: issdandavis/scbe-aethermoore-training-data
233K multi-view SFT pairs (current version)

Author

Issac Davis | ORCID: 0009-0002-3936-9369