Text Generation
Transformers
PyTorch
English
experimental
research
bit-level
transformer
reversible
safety
telemetry
language-modeling
Instructions to use WCNegentropy/BitTransformerLM with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use WCNegentropy/BitTransformerLM with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="WCNegentropy/BitTransformerLM")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("WCNegentropy/BitTransformerLM", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use WCNegentropy/BitTransformerLM with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "WCNegentropy/BitTransformerLM" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "WCNegentropy/BitTransformerLM", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/WCNegentropy/BitTransformerLM
- SGLang
How to use WCNegentropy/BitTransformerLM 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 "WCNegentropy/BitTransformerLM" \ --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": "WCNegentropy/BitTransformerLM", "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 "WCNegentropy/BitTransformerLM" \ --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": "WCNegentropy/BitTransformerLM", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use WCNegentropy/BitTransformerLM with Docker Model Runner:
docker model run hf.co/WCNegentropy/BitTransformerLM
| # BitTransformerLM v0.1.0 - Experimental Research Release | |
| **Release Date:** August 2025 | |
| **Status:** Open Source Research Implementation | |
| **License:** AGPLv3 + Commercial Licensing Available | |
| ## What's Included | |
| This release provides a complete experimental framework for bit-native language modeling research: | |
| - **Core Architecture:** 57 Python files implementing bit-native transformer with reversible layers | |
| - **Safety Systems:** Real-time K/C/S telemetry and monitoring | |
| - **Research Tools:** Interactive dashboard, distributed training, comprehensive testing | |
| - **Documentation:** Professional model card, research status, and validation reports | |
| ## Important Notes | |
| ⚠️ **Experimental Status:** This is research code requiring rigorous baseline validation | |
| ⚠️ **Not Production Ready:** Needs extensive evaluation vs standard transformers | |
| ⚠️ **Research Use Only:** Intended for academic investigation and experimentation | |
| ## Licensing | |
| - **Open Source:** AGPLv3 for research and open source use | |
| - **Commercial:** Contact contact@wcnegentropy.com for commercial licensing | |
| ## Next Steps | |
| The research community is invited to: | |
| 1. Conduct rigorous baseline comparisons vs standard transformers | |
| 2. Evaluate on established language modeling benchmarks | |
| 3. Validate (or refute) claimed memory efficiency benefits | |
| 4. Share findings openly to advance the field | |
| **Research responsibly. Validate rigorously. Share openly.** | |