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

Libraries

How to use shafire/CryptoAI with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("shafire/CryptoAI")
model = AutoModelForCausalLM.from_pretrained("shafire/CryptoAI")
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]:]))

PEFT
How to use shafire/CryptoAI with PEFT:
```
Task type is invalid.
```
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use shafire/CryptoAI with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/shafire/CryptoAI

SGLang

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

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

shafire commited on May 21, 2025

Commit

633e5a8

verified ·

1 Parent(s): 84b0060

Update README.md

Browse files

Files changed (1) hide show

README.md +41 -19

README.md CHANGED Viewed

@@ -1,46 +1,68 @@
 ---
 tags:
 - autotrain
-- text-generation-inference
 - text-generation
 - peft
 library_name: transformers
 base_model: meta-llama/Llama-3.1-8B
-widget:
-  - messages:
-      - role: user
-        content: What is your favorite condiment?
 license: other
 ---
-# Model Trained Using AutoTrain
-This model was trained using AutoTrain. For more information, please visit [AutoTrain](https://hf.co/docs/autotrain).
-# Usage
-```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
-model_path = "PATH_TO_THIS_REPO"
 tokenizer = AutoTokenizer.from_pretrained(model_path)
 model = AutoModelForCausalLM.from_pretrained(
     model_path,
     device_map="auto",
-    torch_dtype='auto'
 ).eval()
-# Prompt content: "hi"
-messages = [
-    {"role": "user", "content": "hi"}
-]
-input_ids = tokenizer.apply_chat_template(conversation=messages, tokenize=True, add_generation_prompt=True, return_tensors='pt')
-output_ids = model.generate(input_ids.to('cuda'))
 response = tokenizer.decode(output_ids[0][input_ids.shape[1]:], skip_special_tokens=True)
-# Model response: "Hello! How can I assist you today?"
 print(response)
-```

 ---
 tags:
 - autotrain
 - text-generation
+- text-generation-inference
 - peft
+- llama-3
+- finance
+- crypto
+- agents
+- workflow-automation
+- soul-ai
 library_name: transformers
 base_model: meta-llama/Llama-3.1-8B
 license: other
+widget:
+  - text: "Ask me something about AI agents or crypto."
+  - text: "What kind of automation can LLMs perform?"
 ---
+# 🧠 $SOUL AI — Llama 3.1 Fine-Tuned for Finance & Autonomous Agents
+**$SOUL AI** is a purpose-tuned LLM based on Meta's Llama 3.1–8B, trained on domain-specific data focused on **financial logic**, **LLM agent workflows**, and **automated task generation**. Designed to power on-chain AI agents, it's part of the broader $SOUL ecosystem for monetized intelligence.
+---
+## 📂 Dataset Summary
+This model was fine-tuned on over 10,000+ instruction-style samples simulating:
+- Financial queries and tokenomics reasoning
+- LLM-agent interaction patterns
+- Crypto automation logic
+- DeFi, trading signals, news interpretation
+- Smart contract and API-triggered tasks
+- Natural language prompts for dynamic workflow creation
+The format follows a custom instruction-based structure optimized for reasoning tasks and agentic workflows—not just casual conversation.
+---
+## 💻 Usage (via Transformers)
+```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
+model_path = "YOUR_HF_USERNAME/YOUR_MODEL_NAME"
 tokenizer = AutoTokenizer.from_pretrained(model_path)
 model = AutoModelForCausalLM.from_pretrained(
     model_path,
     device_map="auto",
+    torch_dtype="auto"
 ).eval()
+messages = [{"role": "user", "content": "How do autonomous LLM agents work?"}]
+input_ids = tokenizer.apply_chat_template(
+    conversation=messages,
+    tokenize=True,
+    add_generation_prompt=True,
+    return_tensors="pt"
+)
+output_ids = model.generate(input_ids.to("cuda"), max_new_tokens=256)
 response = tokenizer.decode(output_ids[0][input_ids.shape[1]:], skip_special_tokens=True)
 print(response)