Instructions to use mrkwanzaa/functionalizer-100M-github-code-python-seed3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mrkwanzaa/functionalizer-100M-github-code-python-seed3 with Transformers:

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

pipe = pipeline("text-generation", model="mrkwanzaa/functionalizer-100M-github-code-python-seed3")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mrkwanzaa/functionalizer-100M-github-code-python-seed3")
model = AutoModelForCausalLM.from_pretrained("mrkwanzaa/functionalizer-100M-github-code-python-seed3", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mrkwanzaa/functionalizer-100M-github-code-python-seed3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mrkwanzaa/functionalizer-100M-github-code-python-seed3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mrkwanzaa/functionalizer-100M-github-code-python-seed3",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/mrkwanzaa/functionalizer-100M-github-code-python-seed3

SGLang

How to use mrkwanzaa/functionalizer-100M-github-code-python-seed3 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 "mrkwanzaa/functionalizer-100M-github-code-python-seed3" \
    --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": "mrkwanzaa/functionalizer-100M-github-code-python-seed3",
		"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 "mrkwanzaa/functionalizer-100M-github-code-python-seed3" \
        --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": "mrkwanzaa/functionalizer-100M-github-code-python-seed3",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use mrkwanzaa/functionalizer-100M-github-code-python-seed3 with Docker Model Runner:
```
docker model run hf.co/mrkwanzaa/functionalizer-100M-github-code-python-seed3
```

Fix metadata: remove arxiv_id, add pipeline tag and library name

by nielsr HF Staff - opened 1 day ago

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

-4

Files changed (1) hide show

README.md +8 -4

README.md CHANGED Viewed

@@ -1,12 +1,16 @@
 ---
-arxiv: 2609.15991
 tags:
-  - functionalizer
-  - tokenizer
-  - gpt2
 ---
 ## Functionalizer 100M model
 This is a 100M example model. Each model is trained on the dataset specified in the name for 50000 epochs with the given seed.
 The model was trained using the Functionalizer framework. Training code and detailed performance analysis are available: https://github.com/connor-makowski/functionalizer

 ---
+library_name: transformers
+pipeline_tag: text-generation
 tags:
+- functionalizer
+- tokenizer
+- gpt2
 ---
 ## Functionalizer 100M model
+This model is based on [The Functionalizer: Lossless Functional Decomposition for Subword Tokenization](https://huggingface.co/papers/2609.15991).
 This is a 100M example model. Each model is trained on the dataset specified in the name for 50000 epochs with the given seed.
 The model was trained using the Functionalizer framework. Training code and detailed performance analysis are available: https://github.com/connor-makowski/functionalizer