OLMo: Open Language Model

OLMo is a repository for training and using AI2's state-of-the-art open language models. It is designed by scientists, for scientists.

Installation

First, install PyTorch following the instructions specific to your operating system.

For training and fine-tuning, we recommend installing from source:

git clone https://github.com/allenai/OLMo.git
cd OLMo
pip install -e .[all]

You can also install from PyPI with:

pip install ai2-olmo

Pretraining

OLMo pretraining follows a two-stage training procedure. In the first stage, we train on large amounts of mostly web-based data: OLMo-mix-1124 In the second stage, we train on a smaller amount of high-quality, targeted data: Dolmino-mix-1124

You can find all the checkpoints, at minimum every 1000 training steps in OLMo core and Hugging Face format:

Variant	OLMo Format	Hugging Face Format
OLMo 7B	OLMo 7B	Hugging Face for the 7B variant
OLMo 13B	OLMo 13B	Hugging Face for the 13B variant

Steps to reproduce

To reproduce any of the training processes described below, run this:

torchrun --nproc_per_node=8 scripts/train.py {path_to_train_config}

For the training config, use any of the configs listed below.

If you want to override any of the settings in the training config without having to write a new config every time, you can do this:

torchrun --nproc_per_node=8 scripts/train.py {path_to_train_config} \
  --setting1=value \
  --setting2=value \
  --setting3.subsetting1=value

The training configs below refer to training data that gets streamed in live over HTTP. To reproduce at large scale, we recommend downloading the files locally and changing the paths to point to your local file system.

Note: Some of the files that the training configs refer to are still being uploaded (as of 2024-11-27). They should all appear in the next few days as the uploads complete.

Stage 1

Stage 1 is the biggest stage, where we train on 4T or 5T tokens on largely web-based data.

	OLMo2 7B	OLMo2 13B
Number of tokens	4 Trillion	5 Trillion
Checkpoint	stage1-step928646-tokens3896B	stage1-step596057-tokens5001B
Training config	OLMo2-7B-stage1.yaml	OLMo2-13B-stage1.yaml
WandB	wandb.ai/OLMo2-7B	wandb.ai/OLMo2-13B

Stage 2 for the 7B

For the 7B model, we train three times with different data order on 50B high quality tokens, and then average ("soup") the models.

	Checkpoint	Training config	WandB
random seed 42	stage2-ingredient1-step11931-tokens50B	OLMo2-7B-stage2-seed42.yaml	wandb.ai/OLMo2-7B
random seed 42069	stage2-ingredient2-step11931-tokens50B	OLMo2-7B-stage2-seed42069.yaml	wandb.ai/OLMo2-7B
random seed 666	stage2-ingredient3-step11931-tokens50B	OLMo2-7B-stage2-seed666.yaml	wandb.ai/OLMo2-7B
final souped model	main	no config, we just averaged the weights in Python

The training configs linked here are set up to download the latest checkpoint after stage 1, and start training from there.

Stage 2 for the 13B

For the 13B model, we train three times with different data order on 100B high quality tokens, and one more time on 300B high quality tokens. Then we average ("soup") the models.

	Checkpoint	Training config	WandB
random seed 1110, 100B	stage2-ingredient1-step11931-tokens100B	OLMo2-13B-stage2-seed1110-100B.yaml	wandb.ai/OLMo2-13B
random seed 2662, 100B	stage2-ingredient2-step11931-tokens100B	OLMo2-13B-stage2-seed2662-100B.yaml	wandb.ai/OLMo2-13B
random seed 6209, 100B	stage2-ingredient3-step11931-tokens100B	OLMo2-13B-stage2-seed6209-100B.yaml	wandb.ai/OLMo2-13B
random seed 2662, 300B	stage2-ingredient4-step11931-tokens300B	OLMo2-13B-stage2-seed2662-300B.yaml	wandb.ai/OLMo2-13B
final souped model	main	no config, we just averaged the weights in Python

The training configs linked here are set up to download the latest checkpoint after stage 1, and start training from there.

Instruction tuned variants

For instruction tuned variants of these models, go to

• OLMo2 7B Instruct
• OLMo2 13B Instruct

Inference

You can use our Hugging Face integration to run inference on the OLMo Transformers checkpoints:

from transformers import AutoModelForCausalLM, AutoTokenizer
olmo = AutoModelForCausalLM.from_pretrained("allenai/OLMo-2-1124-7B")
tokenizer = AutoTokenizer.from_pretrained("allenai/OLMo-2-1124-7B")
message = ["Language modeling is "]
inputs = tokenizer(message, return_tensors='pt', return_token_type_ids=False)
# optional verifying cuda
# inputs = {k: v.to('cuda') for k,v in inputs.items()}
# olmo = olmo.to('cuda')
response = olmo.generate(**inputs, max_new_tokens=100, do_sample=True, top_k=50, top_p=0.95)
print(tokenizer.batch_decode(response, skip_special_tokens=True)[0])

Alternatively, with the Hugging Face pipeline abstraction:

from transformers import pipeline
olmo_pipe = pipeline("text-generation", model="allenai/OLMo-2-1124-7B")
print(olmo_pipe("Language modeling is"))

Quantization

olmo = AutoModelForCausalLM.from_pretrained("allenai/OLMo-2-1124-7B", torch_dtype=torch.float16, load_in_8bit=True)  # requires bitsandbytes

The quantized model is sensitive to input types and CUDA handling. To avoid potential issues, we recommend explicitly converting input IDs to CUDA using: inputs.input_ids.to('cuda')

Evaluation

Additional tools for evaluating OLMo models are available at the OLMo Eval and olmes repositories.

Modal.com Hosting

An example script is provided for hosting an OLMo 2 model on Modal.com using the OpenAI API in ./scripts/olmo2_modal_openai.py. To run that:

1. Follow the instructions under Getting Started in the Modal.com Guide to install the Modal library and command line tools.
2. Follow the instructions under Secrets in the Modal.com Guide to create a Modal secret named "example-secret-token" that defines a value for the variable MODAL_TOKEN for your server.
3. Then run

modal deploy ./scripts/olmo2_modal_openai.py

You can check your endpoint using curl similar to the following:

curl -X POST \
  -H "Authorization: Bearer [the secret token from above]" \
  -H "Content-Type: application/json" \
  -d @body.json \
  https://[the web endpoint modal creates above]/v1/chat/completions

where body.json is of the form:

{
    "model": "OLMo-2-1124-13B-Instruct",
    "messages": [
        {
            "role": "user",
            "content": "Who was Alan Turing?"
        }
      ],
    "max_tokens": 100,
    "temperature": 0.9,
    "stream": true
}

Citing

@misc{olmo20242olmo2furious,
      title={2 OLMo 2 Furious}, 
      author={Team OLMo and Pete Walsh and Luca Soldaini and Dirk Groeneveld and Kyle Lo and Shane Arora and Akshita Bhagia and Yuling Gu and Shengyi Huang and Matt Jordan and Nathan Lambert and Dustin Schwenk and Oyvind Tafjord and Taira Anderson and David Atkinson and Faeze Brahman and Christopher Clark and Pradeep Dasigi and Nouha Dziri and Michal Guerquin and Hamish Ivison and Pang Wei Koh and Jiacheng Liu and Saumya Malik and William Merrill and Lester James V. Miranda and Jacob Morrison and Tyler Murray and Crystal Nam and Valentina Pyatkin and Aman Rangapur and Michael Schmitz and Sam Skjonsberg and David Wadden and Christopher Wilhelm and Michael Wilson and Luke Zettlemoyer and Ali Farhadi and Noah A. Smith and Hannaneh Hajishirzi},
      year={2024},
      eprint={2501.00656},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2501.00656}, 
}