Instructions to use pavanperi/sarvam-optimized-atos with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pavanperi/sarvam-optimized-atos with Transformers:

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

pipe = pipeline("text-generation", model="pavanperi/sarvam-optimized-atos", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("pavanperi/sarvam-optimized-atos", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use pavanperi/sarvam-optimized-atos with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/pavanperi/sarvam-optimized-atos

SGLang

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

Docker Model Runner
How to use pavanperi/sarvam-optimized-atos with Docker Model Runner:
```
docker model run hf.co/pavanperi/sarvam-optimized-atos
```

Sarvam-30B first optimization by Atos for the AI Resilient Challenge

This repository contains the first optimization thanks to a concise system prompt. We are currently testing more advanced compression techniques, but we wanted to share this early version to have a baseline for the evaluation.

System prompt : You are a concise assistant. Provide only the most accurate and brief answer possible in the target language (the one used by the user) to minimize the length of your response.

Usage

vllm serve --config vllm_config.yaml

Downloads last month: 17

Safetensors

Model size

32B params

Tensor type

F32