Instructions to use pre-to-post-olmo/Math-Models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pre-to-post-olmo/Math-Models with Transformers:

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

pipe = pipeline("text-generation", model="pre-to-post-olmo/Math-Models")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("pre-to-post-olmo/Math-Models", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use pre-to-post-olmo/Math-Models with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "pre-to-post-olmo/Math-Models"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "pre-to-post-olmo/Math-Models",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/pre-to-post-olmo/Math-Models

SGLang

How to use pre-to-post-olmo/Math-Models 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 "pre-to-post-olmo/Math-Models" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "pre-to-post-olmo/Math-Models",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "pre-to-post-olmo/Math-Models" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "pre-to-post-olmo/Math-Models",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use pre-to-post-olmo/Math-Models with Docker Model Runner:
```
docker model run hf.co/pre-to-post-olmo/Math-Models
```

Evangelinejy commited on Sep 28

Commit

7afcdd8

verified ·

1 Parent(s): a031ab1

Update README.md

Browse files

Files changed (1) hide show

README.md +15 -0

README.md CHANGED Viewed

@@ -84,3 +84,18 @@ python generate.py --model step95368/rl/step3000 --prompt "..." --max-new-tokens
 `pretrain/` and `anneal/` are base LMs (no chat template); `sft/` and `rl/` are
 chat-tuned and expect the chat template (applied automatically by `generate.py`).

 `pretrain/` and `anneal/` are base LMs (no chat template); `sft/` and `rl/` are
 chat-tuned and expect the chat template (applied automatically by `generate.py`).
+## Citation
+If you use these models, please cite it:
+```bibtex
+@article{pre2post-chess,
+  title   = {Understanding Reasoning from Pretraining to Post-Training},
+  author  = {Shen, Jingyan and Li, Ang and Rahman, Salman and Sun, Yifan and
+             Goldblum, Micah and Telgarsky, Matus and Izmailov, Pavel},
+  journal = {arXiv preprint arXiv:2607.16097},
+  year    = {2026},
+  url     = {https://arxiv.org/pdf/2607.16097}
+}
+```