Acknowledgements¶
LLMForge builds upon the work of many open-source projects and contributors. We thank everyone who has helped make this project possible.
Open Source Projects¶
LLMForge leverages and is inspired by:
- PyTorch - Deep learning framework
- Hugging Face Transformers - Model architecture definitions and tokenizers
- MLX - Apple ML framework
- MLX-LM Community - For the foundational inference code and model implementations
- vLLM - High-performance LLM inference
- Text Generation Inference (TGI) - Hugging Face's inference server
- bitsandbytes - Quantization utilities
- AWQ/GPTQ - Advanced quantization methods
Model Architectures¶
We support numerous model architectures from:
- Meta (Llama)
- Qwen (Alibaba)
- Mistral AI
- Google (Gemma)
- Microsoft (Phi)
- DeepSeek
- And many others...
Contributors¶
Thank you to all contributors who have helped improve LLMForge:
git log --format="%aN" | sort -u
Funding & Support¶
LLMForge is developed by ZandrixAI with support from:
- Open source community
- Beta testers
- Research collaborations
Cross-Platform Support¶
LLMForge supports multiple backends:
- NVIDIA CUDA - via PyTorch
- Apple Metal (MPS/MLX) - Apple Silicon optimization
- Google TPU - via PyTorch/XLA
- CPU - via PyTorch
If we've missed your project or contribution, please let us know!