Instructions to use obiwit/llama3.2-3b-orpo-vanilla-1e with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use obiwit/llama3.2-3b-orpo-vanilla-1e with Transformers:

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

pipe = pipeline("text-generation", model="obiwit/llama3.2-3b-orpo-vanilla-1e", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("obiwit/llama3.2-3b-orpo-vanilla-1e")
model = AutoModelForCausalLM.from_pretrained("obiwit/llama3.2-3b-orpo-vanilla-1e", 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 obiwit/llama3.2-3b-orpo-vanilla-1e with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "obiwit/llama3.2-3b-orpo-vanilla-1e"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "obiwit/llama3.2-3b-orpo-vanilla-1e",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/obiwit/llama3.2-3b-orpo-vanilla-1e

SGLang

How to use obiwit/llama3.2-3b-orpo-vanilla-1e 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 "obiwit/llama3.2-3b-orpo-vanilla-1e" \
    --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": "obiwit/llama3.2-3b-orpo-vanilla-1e",
		"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 "obiwit/llama3.2-3b-orpo-vanilla-1e" \
        --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": "obiwit/llama3.2-3b-orpo-vanilla-1e",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use obiwit/llama3.2-3b-orpo-vanilla-1e with Docker Model Runner:
```
docker model run hf.co/obiwit/llama3.2-3b-orpo-vanilla-1e
```

obiwit commited on Jun 26, 2025

Commit

122b490

verified ·

1 Parent(s): ffaa77f

End of training

Browse files

Files changed (3) hide show

all_results.json +3 -3
config.json +1 -1
eval_results.json +3 -3

all_results.json CHANGED Viewed

@@ -12,10 +12,10 @@
     "eval_rewards/chosen": -0.07526461780071259,
     "eval_rewards/margins": 0.02905045822262764,
     "eval_rewards/rejected": -0.10431348532438278,
-    "eval_runtime": 2433.1431,
     "eval_samples": 189840,
-    "eval_samples_per_second": 78.014,
-    "eval_steps_per_second": 1.219,
     "total_flos": 0.0,
     "train_loss": 0.0,
     "train_runtime": 23.1532,

     "eval_rewards/chosen": -0.07526461780071259,
     "eval_rewards/margins": 0.02905045822262764,
     "eval_rewards/rejected": -0.10431348532438278,
+    "eval_runtime": 2395.4247,
     "eval_samples": 189840,
+    "eval_samples_per_second": 79.242,
+    "eval_steps_per_second": 1.238,
     "total_flos": 0.0,
     "train_loss": 0.0,
     "train_runtime": 23.1532,

config.json CHANGED Viewed

@@ -31,6 +31,6 @@
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.46.3",
-  "use_cache": false,
   "vocab_size": 128256
 }

   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.46.3",
+  "use_cache": true,
   "vocab_size": 128256
 }

eval_results.json CHANGED Viewed

@@ -12,8 +12,8 @@
     "eval_rewards/chosen": -0.07526461780071259,
     "eval_rewards/margins": 0.02905045822262764,
     "eval_rewards/rejected": -0.10431348532438278,
-    "eval_runtime": 2433.1431,
     "eval_samples": 189840,
-    "eval_samples_per_second": 78.014,
-    "eval_steps_per_second": 1.219
 }

     "eval_rewards/chosen": -0.07526461780071259,
     "eval_rewards/margins": 0.02905045822262764,
     "eval_rewards/rejected": -0.10431348532438278,
+    "eval_runtime": 2395.4247,
     "eval_samples": 189840,
+    "eval_samples_per_second": 79.242,
+    "eval_steps_per_second": 1.238
 }