Instructions to use allenai/FlexOlmo-7x7B-1T with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use allenai/FlexOlmo-7x7B-1T with Transformers:

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

pipe = pipeline("text-generation", model="allenai/FlexOlmo-7x7B-1T")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("allenai/FlexOlmo-7x7B-1T")
model = AutoModelForCausalLM.from_pretrained("allenai/FlexOlmo-7x7B-1T", device_map="auto")
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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use allenai/FlexOlmo-7x7B-1T with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "allenai/FlexOlmo-7x7B-1T"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "allenai/FlexOlmo-7x7B-1T",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/allenai/FlexOlmo-7x7B-1T

SGLang

How to use allenai/FlexOlmo-7x7B-1T 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 "allenai/FlexOlmo-7x7B-1T" \
    --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": "allenai/FlexOlmo-7x7B-1T",
		"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 "allenai/FlexOlmo-7x7B-1T" \
        --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": "allenai/FlexOlmo-7x7B-1T",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use allenai/FlexOlmo-7x7B-1T with Docker Model Runner:
```
docker model run hf.co/allenai/FlexOlmo-7x7B-1T
```

Missing Public Base Model

by tomvoelker - opened Aug 4, 2025

Discussion

tomvoelker

Aug 4, 2025

Hey Weijia Shi and team!

First of all, thank you for the great work on the model and for making everything so publicly available! I just wanted to start experimenting with the model by training a custom expert model. However, it seems that the public backbone model linked in the README is currently listed as private or is missing altogether (https://huggingface.co/allenai/Flex-public-7B-1T).

I think I was able to mitigate this by extracting it from one of the 2x7B models. However, to minimise possible sources of error, it would be great if the model could be added as well.

Thanks in advance, and kind regards,
Tom

sewon

Ai2 org Feb 20

It should be uploaded now at https://huggingface.co/allenai/Flex-public-7B-1T. Thanks for raising it and sorry for the late reply!

sewon changed discussion status to closed Feb 20

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment