Instructions to use pragmaticcs/SignOfFour with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pragmaticcs/SignOfFour with Transformers:

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

pipe = pipeline("text-generation", model="pragmaticcs/SignOfFour")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForCausalLM

processor = AutoProcessor.from_pretrained("pragmaticcs/SignOfFour")
model = AutoModelForCausalLM.from_pretrained("pragmaticcs/SignOfFour", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use pragmaticcs/SignOfFour with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/pragmaticcs/SignOfFour

SGLang

How to use pragmaticcs/SignOfFour 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 "pragmaticcs/SignOfFour" \
    --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": "pragmaticcs/SignOfFour",
		"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 "pragmaticcs/SignOfFour" \
        --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": "pragmaticcs/SignOfFour",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use pragmaticcs/SignOfFour with Docker Model Runner:
```
docker model run hf.co/pragmaticcs/SignOfFour
```

pragmaticcs commited on 2 days ago

Commit

4206ccb

verified ·

1 Parent(s): 332b83f

Update README.md

Browse files

Files changed (1) hide show

README.md +4 -4

README.md CHANGED Viewed

@@ -127,15 +127,15 @@ This keeps input/output projections closer to the base and applies the strongest
 ## Layer-Stratified Policies
-| Parameter Group | Match Substring | Policy | Density ($p$) | Base Scale ($\beta$) |
 | :--- | :--- | :---: | :---: | :---: |
 | **Embeddings / LM head** | `embed_tokens`, `lm_head` | Linear | — | 1.00 |
 | **Norms / biases** | `norm`, `bias`, 1D tensors | Linear | — | 1.00 |
 | **DeltaNet recurrent state** | `a_log`, `dt_bias`, `conv1d` | Linear | — | 1.00 |
 | **MoE router gate** | `mlp.gate.weight`, `block_sparse_moe.gate` | Linear | — | 1.00 |
-| **MoE shared expert** | `shared_expert` | DARE-TIES | 0.70 | 0.60 |
-| **Attention projections** | `attn`, `rotary`, `in_proj`, `out_proj`, `x_proj` | DARE-TIES | 0.75 | 0.60 |
-| **Routed experts (×256)** | `experts`, `mlp` | DARE-TIES | 0.65 | 0.55 |
 - **Router protection:** Gate weights use linear interpolation (~57% base, ~43% donors) rather than DARE to avoid destabilizing expert routing.
 - **DeltaNet stability:** Recurrent state kernels are excluded from DARE to prevent divergence in the linear-attention state space.

 ## Layer-Stratified Policies
+| Parameter Group | Match Substring | Policy | Density (p) | Base Scale (β) |
 | :--- | :--- | :---: | :---: | :---: |
 | **Embeddings / LM head** | `embed_tokens`, `lm_head` | Linear | — | 1.00 |
 | **Norms / biases** | `norm`, `bias`, 1D tensors | Linear | — | 1.00 |
 | **DeltaNet recurrent state** | `a_log`, `dt_bias`, `conv1d` | Linear | — | 1.00 |
 | **MoE router gate** | `mlp.gate.weight`, `block_sparse_moe.gate` | Linear | — | 1.00 |
+| **MoE shared expert** | `shared_expert` | DARE&#8209;TIES | 0.70 | 0.60 |
+| **Attention projections** | `attn`, `rotary`, `in_proj`, `out_proj`, `x_proj` | DARE&#8209;TIES | 0.75 | 0.60 |
+| **Routed experts (×256)** | `experts`, `mlp` | DARE&#8209;TIES | 0.65 | 0.55 |
 - **Router protection:** Gate weights use linear interpolation (~57% base, ~43% donors) rather than DARE to avoid destabilizing expert routing.
 - **DeltaNet stability:** Recurrent state kernels are excluded from DARE to prevent divergence in the linear-attention state space.