Instructions to use nmuendler/OpenThinker-7B-text-early-stop-run2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nmuendler/OpenThinker-7B-text-early-stop-run2 with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("open-thoughts/OpenThinker-7B")
model = PeftModel.from_pretrained(base_model, "nmuendler/OpenThinker-7B-text-early-stop-run2")

Transformers

How to use nmuendler/OpenThinker-7B-text-early-stop-run2 with Transformers:

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

pipe = pipeline("text-generation", model="nmuendler/OpenThinker-7B-text-early-stop-run2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("nmuendler/OpenThinker-7B-text-early-stop-run2", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use nmuendler/OpenThinker-7B-text-early-stop-run2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nmuendler/OpenThinker-7B-text-early-stop-run2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nmuendler/OpenThinker-7B-text-early-stop-run2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/nmuendler/OpenThinker-7B-text-early-stop-run2

SGLang

How to use nmuendler/OpenThinker-7B-text-early-stop-run2 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 "nmuendler/OpenThinker-7B-text-early-stop-run2" \
    --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": "nmuendler/OpenThinker-7B-text-early-stop-run2",
		"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 "nmuendler/OpenThinker-7B-text-early-stop-run2" \
        --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": "nmuendler/OpenThinker-7B-text-early-stop-run2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use nmuendler/OpenThinker-7B-text-early-stop-run2 with Docker Model Runner:
```
docker model run hf.co/nmuendler/OpenThinker-7B-text-early-stop-run2
```

OpenThinker-7B-text-early-stop-run2

File size: 1,158 Bytes

585a9da

{
  "event": "training_done",
  "status": "success",
  "phase": "sft",
  "base_model": "open-thoughts/OpenThinker-7B",
  "train_file": "datasets/TLDR_train_filtered_10k.jsonl",
  "output_dir": "outputs/reasoning_abort_sft/openthinker_7b/text/lr5e-05-run2/adapter",
  "task": "summarization",
  "learning_rate": 5e-05,
  "epochs": 1,
  "batch_size": 2,
  "gradient_accumulation_steps": 8,
  "effective_batch_size": 16,
  "max_length": 3000,
  "target_mode": "standard",
  "full_finetune": false,
  "kl_coefficient": 0.0,
  "kl_temperature": 1.0,
  "kl_mask_mode": "full",
  "last_checkpoint": null,
  "function_started_at": "2026-08-18T08:51:53.397627+00:00",
  "training_started_at": "2026-08-18T08:52:24.704136+00:00",
  "ended_at": "2026-08-18T09:07:51.257658+00:00",
  "wall_clock_seconds": 926.5535331210122,
  "seconds": 926.5535331210122,
  "end_to_end_seconds": 957.8600465520285,
  "trainer_train_runtime": 925.7075,
  "trainer_metrics": {
    "train_runtime": 925.7075,
    "train_samples_per_second": 10.803,
    "train_steps_per_second": 0.675,
    "total_flos": 4173075710189568.0,
    "train_loss": 2.5310071309407554,
    "epoch": 0.0192
  }
}