Instructions to use suratkiade/the-cohesive-tetrad-instruct-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use suratkiade/the-cohesive-tetrad-instruct-base with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="suratkiade/the-cohesive-tetrad-instruct-base")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("suratkiade/the-cohesive-tetrad-instruct-base")
model = AutoModelForCausalLM.from_pretrained("suratkiade/the-cohesive-tetrad-instruct-base")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use suratkiade/the-cohesive-tetrad-instruct-base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "suratkiade/the-cohesive-tetrad-instruct-base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "suratkiade/the-cohesive-tetrad-instruct-base",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/suratkiade/the-cohesive-tetrad-instruct-base

SGLang

How to use suratkiade/the-cohesive-tetrad-instruct-base 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 "suratkiade/the-cohesive-tetrad-instruct-base" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "suratkiade/the-cohesive-tetrad-instruct-base",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "suratkiade/the-cohesive-tetrad-instruct-base" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "suratkiade/the-cohesive-tetrad-instruct-base",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use suratkiade/the-cohesive-tetrad-instruct-base with Docker Model Runner:
```
docker model run hf.co/suratkiade/the-cohesive-tetrad-instruct-base
```

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

The Cohesive Tetrad — Instruct Base (TinyLlama Mirror)

suratkiade/the-cohesive-tetrad-instruct-base adalah canonical base model untuk seluruh varian instruksi The Cohesive Tetrad (TCT) di Hugging Face.

Model ini merupakan mirror bobot 1 : 1 dari:

TinyLlama/TinyLlama-1.1B-Chat-v1.0

dengan lisensi Apache-2.0.

Tidak ada fine-tuning TCT di model ini

Bobot model identik dengan rilis TinyLlama asli; belum mengandung injeksi semantik empat pilar TCT (Sabda, Logic, Qualia, Mystica) di dalam parameter model.
Tidak ada data kanonis TCT yang digunakan pada tahap pelatihan model ini.
Model ini berfungsi sebagai baseline arsitektural sebelum TCT diinjeksikan.

Peran di arsitektur The Cohesive Tetrad

Peran utama: baseline yang stabil, terbuka, dan tidak bergated untuk seluruh varian The Cohesive Tetrad downstream
(misalnya suratkiade/the-cohesive-tetrad-instruct-v1).
Format bobot: BF16 safetensors; kompatibel dengan transformers.
Tipe model: varian TinyLlama chat, sebelum injeksi semantik TCT.

Kapan memakai model ini

Gunakan model …-instruct-base jika:

Anda memerlukan titik awal yang netral untuk uji coba dan eksperimen arsitektural berbasis TinyLlama (misalnya membandingkan sebelum dan sesudah TCT fine-tuning).
Anda ingin menjalankan prosedur fine-tuning sendiri di atas TinyLlama tanpa terlebih dahulu memakai model yang sudah di-fine-tune dengan data TCT.
Anda ingin mereplikasi atau mengaudit proses canonical fine-tuning yang menghasilkan model suratkiade/the-cohesive-tetrad-instruct-v1.

“Untuk berdialog langsung dengan semantik The Cohesive Tetrad yang sudah di-fine-tune, gunakan model suratkiade/the-cohesive-tetrad-instruct-v1 sebagai rilis instruksi kanonis pertama.”

Inference API (requests)

Tanpa instalasi lokal — siapkan token HF:

export HF_TOKEN=hf_...

Downloads last month: -

Safetensors

Model size

1B params

Tensor type

BF16

Model tree for suratkiade/the-cohesive-tetrad-instruct-base

Base model

TinyLlama/TinyLlama-1.1B-Chat-v1.0

Finetuned

(557)

this model

Finetunes

1 model