Skip to main content
Glama

Zignet

by fulgidus
Do What The F*ck You Want To Public License
7
requirements-train.txt1.05 kB
# ZigNet Training Dependencies # For RTX 3090 with CUDA 12.7 # Tested compatible versions (October 2025) # Core dependencies - PINNED versions for compatibility torch==2.4.1 torchvision==0.19.1 torchaudio==2.4.1 # Transformers & Training - Compatible versions transformers==4.45.2 datasets==2.21.0 peft==0.13.2 # LoRA implementation trl==0.11.4 # Supervised Fine-Tuning Trainer accelerate==0.34.2 # Optimization bitsandbytes==0.44.1 # 4-bit/8-bit quantization scipy==1.14.1 # Utilities sentencepiece==0.2.0 # Tokenizer protobuf==3.20.3 psutil==6.1.1 # Unsloth - install last unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git # HuggingFace Hub (optional) huggingface-hub==0.26.5 # Optimization bitsandbytes>=0.44.0 # 4-bit/8-bit quantization scipy>=1.11.0 # Utilities sentencepiece>=0.2.0 # Tokenizer protobuf>=3.20.0 psutil>=5.9.0 # HuggingFace Hub (optional) huggingface-hub>=0.24.0 # Monitoring (optional) # wandb>=0.17.0 # Uncomment for experiment tracking # tensorboard>=2.14.0 # Uncomment for TensorBoard logging

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/fulgidus/zignet'

If you have feedback or need assistance with the MCP directory API, please join our Discord server