Text Generation
Transformers
Safetensors
jirack_ternary
ternary
bitnet
1.58bit
llama
ternary-transformer
quantized
70b
jirack
amd-rocm
vram-optimized
Eval Results (legacy)
2-bit
Instructions to use kgrabko/JiRackTernary_70b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use kgrabko/JiRackTernary_70b with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="kgrabko/JiRackTernary_70b")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("kgrabko/JiRackTernary_70b", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use kgrabko/JiRackTernary_70b with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "kgrabko/JiRackTernary_70b" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "kgrabko/JiRackTernary_70b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/kgrabko/JiRackTernary_70b
- SGLang
How to use kgrabko/JiRackTernary_70b 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 "kgrabko/JiRackTernary_70b" \ --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": "kgrabko/JiRackTernary_70b", "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 "kgrabko/JiRackTernary_70b" \ --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": "kgrabko/JiRackTernary_70b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use kgrabko/JiRackTernary_70b with Docker Model Runner:
docker model run hf.co/kgrabko/JiRackTernary_70b
| # ============================================================================== | |
| # COPYRIGHT (C) 2025-2026 KONSTANTIN VLADIMIROVICH GRABKO. ALL RIGHTS RESERVED. | |
| # PATENT PENDING | CMS MANHATTAN JIRACK TECHNOLOGY | |
| # ============================================================================== | |
| import os | |
| import json | |
| from safetensors import safe_open | |
| MODEL_PATH = "./JiRackTernary_70B_converted" | |
| files = [f for f in os.listdir(MODEL_PATH) if f.endswith(".safetensors")] | |
| files.sort() | |
| metadata = {} | |
| weight_map = {} | |
| print(f"🔍 Сканирую {len(files)} шардов...") | |
| for filename in files: | |
| full_path = os.path.join(MODEL_PATH, filename) | |
| with safe_open(full_path, framework="pt") as f: | |
| for key in f.keys(): | |
| weight_map[key] = filename | |
| index_data = { | |
| "metadata": {"total_size": 0}, # Размер можно не указывать точно | |
| "weight_map": weight_map | |
| } | |
| with open(os.path.join(MODEL_PATH, "model.safetensors.index.json"), "w") as f: | |
| json.dump(index_data, f, indent=2) | |
| print("✅ Файл model.safetensors.index.json успешно создан!") | |