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

Libraries

How to use cortexso/aya with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="cortexso/aya",
	filename="aya-23-35b-q2_k.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use cortexso/aya with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf cortexso/aya:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf cortexso/aya:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf cortexso/aya:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf cortexso/aya:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf cortexso/aya:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf cortexso/aya:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf cortexso/aya:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf cortexso/aya:Q4_K_M

Use Docker

docker model run hf.co/cortexso/aya:Q4_K_M

LM Studio
Jan

vLLM

How to use cortexso/aya with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/cortexso/aya:Q4_K_M

Ollama
How to use cortexso/aya with Ollama:
```
ollama run hf.co/cortexso/aya:Q4_K_M
```

Unsloth Studio new

How to use cortexso/aya 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 cortexso/aya 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 cortexso/aya to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for cortexso/aya to start chatting

Docker Model Runner
How to use cortexso/aya with Docker Model Runner:
```
docker model run hf.co/cortexso/aya:Q4_K_M
```

Lemonade

How to use cortexso/aya with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull cortexso/aya:Q4_K_M

Run and chat with the model

lemonade run user.aya-Q4_K_M

List all available models

lemonade list

Overview

Cohere For AI developed and released the Aya 23, an open weights instruction fine-tuned model with advanced multilingual capabilities. Aya 23 is built upon the highly performant Command family of models and fine-tuned using the Aya Collection to deliver state-of-the-art performance across 23 languages. This multilingual large language model is designed to support a wide range of use cases, including multilingual text generation, understanding, and translation tasks.

Aya 23, balancing efficiency and performance. It offers robust multilingual support for languages such as Arabic, Chinese, English, Spanish, Hindi, Vietnamese, and more, making it a versatile tool for global applications. A 35-billion parameter version is also available here.

Variants

No	Variant	Cortex CLI command
1	Aya-8b	`cortex run aya:8b`
2	Aya-35b	`cortex run aya:35b`

Use it with Jan (UI)

Install Jan using Quickstart
Use in Jan model Hub:
```
cortexso/aya
```

Use it with Cortex (CLI)

Install Cortex using Quickstart
Run the model with command:
```
cortex run aya
```

Credits

Author: Cohere For AI
Converter: Homebrew
Original License: License

Downloads last month: 351

GGUF

Model size

35B params

Architecture

command-r

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit