Text Generation
Transformers
ONNX
Safetensors
English
qwen2
dictation
cleanup
transcript
lora
mumble
conversational
text-generation-inference
Instructions to use adikuma/mumble-cleanup with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
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)# Load model directly from transformers import AutoTokenizer, AutoModelForMultimodalLM tokenizer = AutoTokenizer.from_pretrained("adikuma/mumble-cleanup") model = AutoModelForMultimodalLM.from_pretrained("adikuma/mumble-cleanup") 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=40) 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
| """push the cleanup code + seed dataset to a fresh huggingface hub repo. | |
| does not need a trained model. uploads: | |
| - src/, scripts/, configs/, tests/, data/, docs/ | |
| - pyproject.toml, Makefile | |
| - docs/model_card.md mirrored to README.md at repo root | |
| - existing cleanup/README.md renamed to DEVELOPMENT.md | |
| reads the token from the mumble repo root .env.local. prefers | |
| HUGGINGFACE_ACCESS_TOKEN, falls back to HF_TOKEN. | |
| usage: python models/cleanup/scripts/push_code_to_hub.py [--repo NAME] [--private] | |
| """ | |
| import argparse | |
| import shutil | |
| import sys | |
| import tempfile | |
| from pathlib import Path | |
| from huggingface_hub import HfApi | |
| from huggingface_hub.errors import HfHubHTTPError | |
| SKIP_DIRS = { | |
| ".venv", | |
| "runs", | |
| "dist", | |
| "__pycache__", | |
| ".pytest_cache", | |
| ".mypy_cache", | |
| ".ruff_cache", | |
| "node_modules", | |
| ".uv", | |
| } | |
| SKIP_SUFFIXES = {".pyc", ".pyo"} | |
| def find_repo_root(start: Path) -> Path: | |
| for p in [start, *start.parents]: | |
| if (p / ".env.local").exists(): | |
| return p | |
| raise FileNotFoundError(".env.local not found in any parent of " + str(start)) | |
| def load_token(env_path: Path) -> str: | |
| for raw in env_path.read_text(encoding="utf-8").splitlines(): | |
| line = raw.strip() | |
| if not line or line.startswith("#") or "=" not in line: | |
| continue | |
| key, _, value = line.partition("=") | |
| key = key.strip() | |
| value = value.strip().strip('"').strip("'") | |
| if key in {"HUGGINGFACE_ACCESS_TOKEN", "HF_TOKEN"} and value: | |
| return value | |
| raise KeyError("no HUGGINGFACE_ACCESS_TOKEN or HF_TOKEN in " + str(env_path)) | |
| def stage_upload(source: Path, staging: Path) -> int: | |
| count = 0 | |
| for path in source.rglob("*"): | |
| if path.is_dir(): | |
| continue | |
| rel = path.relative_to(source) | |
| if SKIP_DIRS & set(rel.parts): | |
| continue | |
| if path.suffix in SKIP_SUFFIXES: | |
| continue | |
| if rel == Path("README.md"): | |
| target_rel = Path("DEVELOPMENT.md") | |
| else: | |
| target_rel = rel | |
| target = staging / target_rel | |
| target.parent.mkdir(parents=True, exist_ok=True) | |
| shutil.copy2(path, target) | |
| count += 1 | |
| model_card = source / "docs" / "model_card.md" | |
| if model_card.exists(): | |
| shutil.copy2(model_card, staging / "README.md") | |
| count += 1 | |
| else: | |
| print("warn: docs/model_card.md missing; hub page will lack a README", file=sys.stderr) | |
| (staging / ".gitattributes").write_text( | |
| "*.jsonl text\n*.csv text\n*.md text\n*.py text\n*.yaml text\n*.toml text\n", | |
| encoding="utf-8", | |
| ) | |
| return count + 1 | |
| def main() -> int: | |
| parser = argparse.ArgumentParser() | |
| parser.add_argument("--repo", default="mumble-cleanup", help="repo name under the authed user") | |
| parser.add_argument("--private", action="store_true", help="create as private (default public)") | |
| args = parser.parse_args() | |
| here = Path(__file__).resolve() | |
| repo_root = find_repo_root(here.parent) | |
| cleanup_dir = repo_root / "models" / "cleanup" | |
| env_path = repo_root / ".env.local" | |
| if not cleanup_dir.exists(): | |
| print(f"error: {cleanup_dir} does not exist", file=sys.stderr) | |
| return 1 | |
| token = load_token(env_path) | |
| api = HfApi(token=token) | |
| me = api.whoami() | |
| user = me.get("name") or (me.get("email") or "").split("@")[0] | |
| if not user: | |
| print("error: could not resolve hf username from whoami()", file=sys.stderr) | |
| return 1 | |
| repo_id = f"{user}/{args.repo}" | |
| print(f"hf user : {user}") | |
| print(f"repo target : {repo_id}") | |
| print(f"visibility : {'private' if args.private else 'public'}") | |
| print(f"source : {cleanup_dir}") | |
| try: | |
| url = api.create_repo( | |
| repo_id=repo_id, | |
| private=args.private, | |
| repo_type="model", | |
| exist_ok=True, | |
| ) | |
| print(f"repo ready : {url}") | |
| except HfHubHTTPError as exc: | |
| print(f"error: failed to create repo: {exc}", file=sys.stderr) | |
| return 1 | |
| with tempfile.TemporaryDirectory(prefix="mumble-cleanup-hub-") as tmp: | |
| staging = Path(tmp) | |
| count = stage_upload(cleanup_dir, staging) | |
| print(f"staged files : {count}") | |
| print("uploading ...") | |
| api.upload_folder( | |
| folder_path=str(staging), | |
| repo_id=repo_id, | |
| repo_type="model", | |
| commit_message="initial upload: cleanup code and 688-pair seed dataset", | |
| ) | |
| print(f"\ndone. browse: https://huggingface.co/{repo_id}") | |
| return 0 | |
| if __name__ == "__main__": | |
| raise SystemExit(main()) | |