Instructions to use physicsrob/torchwright-doom-e1m1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use physicsrob/torchwright-doom-e1m1 with Transformers:

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

pipe = pipeline("text-generation", model="physicsrob/torchwright-doom-e1m1")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("physicsrob/torchwright-doom-e1m1")
model = AutoModelForCausalLM.from_pretrained("physicsrob/torchwright-doom-e1m1", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use physicsrob/torchwright-doom-e1m1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "physicsrob/torchwright-doom-e1m1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "physicsrob/torchwright-doom-e1m1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/physicsrob/torchwright-doom-e1m1

SGLang

How to use physicsrob/torchwright-doom-e1m1 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 "physicsrob/torchwright-doom-e1m1" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "physicsrob/torchwright-doom-e1m1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "physicsrob/torchwright-doom-e1m1" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "physicsrob/torchwright-doom-e1m1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use physicsrob/torchwright-doom-e1m1 with Docker Model Runner:
```
docker model run hf.co/physicsrob/torchwright-doom-e1m1
```

torchwright-doom-e1m1

File size: 746 Bytes

b9a8681

# Reproduce the Doom render

Run stock Hugging Face inference first, from the bundle root:

```bash
python infer.py --model . --prompt examples/e1m1_prompt.txt --output out
```

The two following tools independently consume the raw tokenizer text. They
do not load the model and do not invoke one another:

```bash
python tools/pretty_text.py --input out/output.txt --output out/output.pretty.txt
python tools/txt_to_png.py  --input out/output.txt --output out/frame.png
```

`output.ids.json` is the canonical record of emitted model row ids;
`output.txt` is their standard-tokenizer interchange form.

The cursor/pixel token protocol these tools decode is specified in
`PROTOCOL.md` in the source repo (github.com/physicsrob/torchwright_doom).