Instructions to use maym15/Olmo-3-7B-Instruct-RetMask with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use maym15/Olmo-3-7B-Instruct-RetMask with Transformers:

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

pipe = pipeline("text-generation", model="maym15/Olmo-3-7B-Instruct-RetMask")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("maym15/Olmo-3-7B-Instruct-RetMask")
model = AutoModelForCausalLM.from_pretrained("maym15/Olmo-3-7B-Instruct-RetMask", 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use maym15/Olmo-3-7B-Instruct-RetMask with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "maym15/Olmo-3-7B-Instruct-RetMask"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "maym15/Olmo-3-7B-Instruct-RetMask",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/maym15/Olmo-3-7B-Instruct-RetMask

SGLang

How to use maym15/Olmo-3-7B-Instruct-RetMask 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 "maym15/Olmo-3-7B-Instruct-RetMask" \
    --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": "maym15/Olmo-3-7B-Instruct-RetMask",
		"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 "maym15/Olmo-3-7B-Instruct-RetMask" \
        --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": "maym15/Olmo-3-7B-Instruct-RetMask",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use maym15/Olmo-3-7B-Instruct-RetMask with Docker Model Runner:
```
docker model run hf.co/maym15/Olmo-3-7B-Instruct-RetMask
```

maym15 commited on Apr 21

Commit

f30f04f

verified ·

1 Parent(s): 5421f85

Update README.md

Browse files

Files changed (1) hide show

README.md +18 -4

README.md CHANGED Viewed

@@ -1,7 +1,7 @@
 ---
 base_model: allenai/Olmo-3-7B-Instruct
 library_name: transformers
-model_name: olmo-3-instruct-masked_102heads_LR_5e-7_MINLR_5e-8_WD_0.1_42
 tags:
 - generated_from_trainer
 - hf_jobs
@@ -10,7 +10,7 @@ tags:
 licence: license
 ---
-# Model Card for olmo-3-instruct-masked_102heads_LR_5e-7_MINLR_5e-8_WD_0.1_42
 This model is a fine-tuned version of [allenai/Olmo-3-7B-Instruct](https://huggingface.co/allenai/Olmo-3-7B-Instruct).
 It has been trained using [TRL](https://github.com/huggingface/trl).
@@ -28,8 +28,6 @@ print(output["generated_text"])
 ## Training procedure
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/youmima/dpo/runs/l6wg0w7q)
 This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
@@ -43,6 +41,22 @@ This model was trained with DPO, a method introduced in [Direct Preference Optim
 ## Citations
 Cite DPO as:
 ```bibtex

 ---
 base_model: allenai/Olmo-3-7B-Instruct
 library_name: transformers
+model_name: Olmo-3-7B-Instruct-RetMask
 tags:
 - generated_from_trainer
 - hf_jobs
 licence: license
 ---
+# Model Card for Olmo-3-7B-Instruct-RetMask
 This model is a fine-tuned version of [allenai/Olmo-3-7B-Instruct](https://huggingface.co/allenai/Olmo-3-7B-Instruct).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 ## Training procedure
 This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
 ## Citations
+Cite RetMask as:
+```bibtex
+@inproceedings{ma2026interpretability,
+    title     = "From Interpretability to Performance: Optimizing Retrieval Heads for Long-Context Language Models",
+    author    = "Youmi Ma and Naoaki Okazaki",
+    booktitle = "Findings of the Association for Computational Linguistics: ACL 2026",
+    month     = jul,
+    year      = "2026",
+    address   = "San Diego, California, United States",
+    publisher = "Association for Computational Linguistics",
+    pages     = "(TBD)",
+}
+```
 Cite DPO as:
 ```bibtex