Instructions to use adikuma/mumble-cleanup with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use adikuma/mumble-cleanup with Transformers:

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

pipe = pipeline("text-generation", model="adikuma/mumble-cleanup")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("adikuma/mumble-cleanup")
model = AutoModelForCausalLM.from_pretrained("adikuma/mumble-cleanup", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use adikuma/mumble-cleanup with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/adikuma/mumble-cleanup

SGLang

How to use adikuma/mumble-cleanup 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 "adikuma/mumble-cleanup" \
    --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": "adikuma/mumble-cleanup",
		"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 "adikuma/mumble-cleanup" \
        --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": "adikuma/mumble-cleanup",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use adikuma/mumble-cleanup with Docker Model Runner:
```
docker model run hf.co/adikuma/mumble-cleanup
```

mumble-cleanup

File size: 1,762 Bytes

fd0b01f

.PHONY: help sync data train evaluate export benchmark pack push all smoke clean

RUN_ID ?= run-$(shell date -u +%Y%m%d-%H%M%S)
LR ?=
EPOCHS ?=

help:
	@echo "make sync                  install dependencies via uv"
	@echo "make data                  01: download disfluencyspeech, build injection pairs"
	@echo "make train RUN_ID=r1       02: lora sft on qwen2.5-0.5b-instruct"
	@echo "make evaluate RUN_ID=r1    03: raw vs base vs fine-tune on the real held-out test"
	@echo "make export RUN_ID=r1      04: merge lora and export fp32 + int8 onnx"
	@echo "make benchmark RUN_ID=r1   05: cpu latency (run locally for the truth)"
	@echo "make pack RUN_ID=r1        06: tar the run, print sha256 and the scp line"
	@echo "make push RUN_ID=r1        publish to the hf hub (uses HF_TOKEN in .env.local)"
	@echo "make all RUN_ID=r1         the whole pipeline end to end (no push, no benchmark)"
	@echo "make smoke                 tiny cpu run, no gpu needed, validates wiring"

sync:
	uv sync

data:
	uv run python scripts/01_download.py

train:
	uv run python scripts/02_train.py --run-id $(RUN_ID) $(if $(LR),--lr $(LR),) $(if $(EPOCHS),--epochs $(EPOCHS),)

evaluate:
	uv run python scripts/03_evaluate.py --run-id $(RUN_ID)

export:
	uv run python scripts/04_export.py --run-id $(RUN_ID)

benchmark:
	uv run python scripts/05_benchmark.py --run-id $(RUN_ID)

pack:
	uv run python scripts/06_pack_and_ship.py --run-id $(RUN_ID)

push:
	uv run python scripts/push_to_hub.py --run-id $(RUN_ID)

all: sync data train evaluate export pack

smoke:
	uv run python scripts/01_download.py --smoke
	uv run python scripts/02_train.py --run-id smoke --smoke
	uv run python scripts/03_evaluate.py --run-id smoke --smoke

clean:
	rm -rf runs/*
	rm -rf data/pairs
	rm -rf dist/*