Text Generation
Transformers
Safetensors
English
Russian
knk_vf
nullxes
knkf
knk-vf
void-forged
kurotama-no-kami
Mixture of Experts
sparse-moe
mixture-of-experts
initialization
random-init
bf16
long-context
multilingual
code
enterprise
h200
b300
megatron
Instructions to use MagistrTheOne/KNK-VF-Lab-38B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use MagistrTheOne/KNK-VF-Lab-38B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="MagistrTheOne/KNK-VF-Lab-38B")# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("MagistrTheOne/KNK-VF-Lab-38B", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use MagistrTheOne/KNK-VF-Lab-38B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "MagistrTheOne/KNK-VF-Lab-38B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "MagistrTheOne/KNK-VF-Lab-38B", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/MagistrTheOne/KNK-VF-Lab-38B
- SGLang
How to use MagistrTheOne/KNK-VF-Lab-38B 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 "MagistrTheOne/KNK-VF-Lab-38B" \ --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": "MagistrTheOne/KNK-VF-Lab-38B", "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 "MagistrTheOne/KNK-VF-Lab-38B" \ --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": "MagistrTheOne/KNK-VF-Lab-38B", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use MagistrTheOne/KNK-VF-Lab-38B with Docker Model Runner:
docker model run hf.co/MagistrTheOne/KNK-VF-Lab-38B
| { | |
| "config_path": "configs/model/knk_vf_lab_38b_active5b.yaml", | |
| "output_dir": "/workspace/checkpoints/knk_vf_lab_38b", | |
| "shards": 9, | |
| "tensors": 7311, | |
| "total_params": 38049091584, | |
| "active_params": 5349249024, | |
| "total_bytes": 76098763776, | |
| "dtype": "bfloat16", | |
| "device": "cpu" | |
| } |