Instructions to use cortexso/olmo-2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cortexso/olmo-2 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="cortexso/olmo-2",
	filename="olmo-2-0325-32b-instruct-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/olmo-2 with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf cortexso/olmo-2:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf cortexso/olmo-2: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/olmo-2:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf cortexso/olmo-2: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/olmo-2:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf cortexso/olmo-2: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/olmo-2:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf cortexso/olmo-2:Q4_K_M

Use Docker

docker model run hf.co/cortexso/olmo-2:Q4_K_M

LM Studio
Jan

vLLM

How to use cortexso/olmo-2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cortexso/olmo-2"
# 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/olmo-2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cortexso/olmo-2:Q4_K_M

Ollama
How to use cortexso/olmo-2 with Ollama:
```
ollama run hf.co/cortexso/olmo-2:Q4_K_M
```

Unsloth Studio new

How to use cortexso/olmo-2 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/olmo-2 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/olmo-2 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for cortexso/olmo-2 to start chatting

Docker Model Runner
How to use cortexso/olmo-2 with Docker Model Runner:
```
docker model run hf.co/cortexso/olmo-2:Q4_K_M
```

Lemonade

How to use cortexso/olmo-2 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull cortexso/olmo-2:Q4_K_M

Run and chat with the model

lemonade run user.olmo-2-Q4_K_M

List all available models

lemonade list

Overview

OLMo-2 is a series of Open Language Models designed to enable the science of language models. These models are trained on the Dolma dataset, with all code, checkpoints, logs (coming soon), and associated training details made openly available.

The OLMo-2 13B Instruct November 2024 is a post-trained variant of the OLMo-2 13B model, which has undergone supervised fine-tuning on an OLMo-specific variant of the Tülu 3 dataset. Additional training techniques include Direct Preference Optimization (DPO) and Reinforcement Learning from Virtual Rewards (RLVR), optimizing it for state-of-the-art performance across various tasks, including chat, MATH, GSM8K, and IFEval.

Variants

No	Variant	Cortex CLI command
1	Olmo-2-7b	`cortex run olmo-2:7b`
2	Olmo-2-13b	`cortex run olmo-2:13b`
3	Olmo-2-32b	`cortex run olmo-2:32b`

Use it with Jan (UI)

Install Jan using Quickstart
Use in Jan model Hub:
```
cortexhub/olmo-2
```

Use it with Cortex (CLI)

Install Cortex using Quickstart
Run the model with command:
```
cortex run olmo-2
```

Credits

Author: allenai
Converter: Homebrew
Original License: Licence
Papers: Paper

Downloads last month: 321

GGUF

Model size

32B params

Architecture

olmo2

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Paper for cortexso/olmo-2

2 OLMo 2 Furious

Paper • 2501.00656 • Published Dec 31, 2024 • 22