Instructions to use aayanmishra-ml/AwA-1.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use aayanmishra-ml/AwA-1.5B with Transformers:

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

pipe = pipeline("text-generation", model="aayanmishra-ml/AwA-1.5B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("aayanmishra-ml/AwA-1.5B")
model = AutoModelForCausalLM.from_pretrained("aayanmishra-ml/AwA-1.5B")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use aayanmishra-ml/AwA-1.5B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "aayanmishra-ml/AwA-1.5B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "aayanmishra-ml/AwA-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/aayanmishra-ml/AwA-1.5B

SGLang

How to use aayanmishra-ml/AwA-1.5B 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 "aayanmishra-ml/AwA-1.5B" \
    --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": "aayanmishra-ml/AwA-1.5B",
		"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 "aayanmishra-ml/AwA-1.5B" \
        --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": "aayanmishra-ml/AwA-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio new

How to use aayanmishra-ml/AwA-1.5B with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for aayanmishra-ml/AwA-1.5B to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for aayanmishra-ml/AwA-1.5B to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for aayanmishra-ml/AwA-1.5B to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="aayanmishra-ml/AwA-1.5B",
    max_seq_length=2048,
)

Docker Model Runner
How to use aayanmishra-ml/AwA-1.5B with Docker Model Runner:
```
docker model run hf.co/aayanmishra-ml/AwA-1.5B
```

Aayan Mishra commited on Dec 29, 2024

Commit

46aa307

verified ·

1 Parent(s): b98e873

Update README.md

Browse files

Files changed (1) hide show

README.md +68 -6

README.md CHANGED Viewed

@@ -9,14 +9,76 @@ tags:
 license: apache-2.0
 language:
 - en
 ---
-# Uploaded  model
-- **Developed by:** Spestly
-- **License:** apache-2.0
-- **Finetuned from model :** Spestly/Athena-2-1.5B
-This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
-[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)

 license: apache-2.0
 language:
 - en
+library_name: transformers
 ---
+![Header](https://raw.githubusercontent.com/Aayan-Mishra/Images/refs/heads/main/AwA.png)
+# AwA - 1.5B
+AwA (Answers with Athena) is my portfolio project, showcasing a cutting-edge Chain-of-Thought (CoT) reasoning model. I created AwA to excel in providing detailed, step-by-step answers to complex questions across diverse domains. This model represents my dedication to advancing AI’s capability for enhanced comprehension, problem-solving, and knowledge synthesis.
+## Key Features
+- **Chain-of-Thought Reasoning:** AwA delivers step-by-step breakdowns of solutions, mimicking logical human thought processes.
+- **Domain Versatility:** Performs exceptionally across a wide range of domains, including mathematics, science, literature, and more.
+- **Adaptive Responses:** Adjusts answer depth and complexity based on input queries, catering to both novices and experts.
+- **Interactive Design:** Designed for educational tools, research assistants, and decision-making systems.
+## Intended Use Cases
+- **Educational Applications:** Supports learning by breaking down complex problems into manageable steps.
+- **Research Assistance:** Generates structured insights and explanations in academic or professional research.
+- **Decision Support:** Enhances understanding in business, engineering, and scientific contexts.
+- **General Inquiry:** Provides coherent, in-depth answers to everyday questions.
+# Type: Chain-of-Thought (CoT) Reasoning Model
+- Base Architecture: Adapted from [qwen2]
+- Parameters: [1.54B]
+- Fine-tuning: Specialized fine-tuning on Chain-of-Thought reasoning datasets to enhance step-by-step explanatory capabilities.
+## Ethical Considerations
+- **Bias Mitigation:** I have taken steps to minimise biases in the training data. However, users are encouraged to cross-verify outputs in sensitive contexts.
+- **Limitations:** May not provide exhaustive answers for niche topics or domains outside its training scope.
+- **User Responsibility:** Designed as an assistive tool, not a replacement for expert human judgment.
+## Usage
+### Option A: Local
+Using locally with the Transformers library
+```python
+# Use a pipeline as a high-level helper
+from transformers import pipeline
+messages = [
+    {"role": "user", "content": "Who are you?"},
+]
+pipe = pipeline("text-generation", model="Spestly/AwA-1.5B")
+pipe(messages)
+```
+### Option B: API & Space
+You can use the AwA HuggingFace space or the AwA API (Coming soon!)
+## Roadmap
+- More AwA model sizes e.g 7B and 14B
+- Create AwA API via spestly package