Instructions to use muradil211/AetherSearch_DPO with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use muradil211/AetherSearch_DPO with Transformers:

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

pipe = pipeline("text-generation", model="muradil211/AetherSearch_DPO")
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("muradil211/AetherSearch_DPO")
model = AutoModelForCausalLM.from_pretrained("muradil211/AetherSearch_DPO", 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 muradil211/AetherSearch_DPO with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "muradil211/AetherSearch_DPO"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "muradil211/AetherSearch_DPO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/muradil211/AetherSearch_DPO

SGLang

How to use muradil211/AetherSearch_DPO 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 "muradil211/AetherSearch_DPO" \
    --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": "muradil211/AetherSearch_DPO",
		"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 "muradil211/AetherSearch_DPO" \
        --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": "muradil211/AetherSearch_DPO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use muradil211/AetherSearch_DPO with Docker Model Runner:
```
docker model run hf.co/muradil211/AetherSearch_DPO
```

muradil211 commited on Aug 31

Commit

f5edce8

verified ·

1 Parent(s): 6eff564

Fix inline math rendering in model card

Browse files

Use the Hugging Face-supported \$...\$ delimiters for inline KaTeX while retaining $$...$$ for the display DPO objective.

Files changed (1) hide show

README.md +4 -4

README.md CHANGED Viewed

@@ -93,9 +93,9 @@ result as `<information>`, and resumes generation until the model emits an
 Supervised fine-tuning teaches the model how to follow the search protocol;
 DPO then teaches it which of two valid-looking continuations is preferable.
-For a shared prompt $x$, preferred continuation $y_w$, rejected continuation
-$y_l$, policy $\pi_\theta$, and frozen SFT reference $\pi_{\mathrm{ref}}$,
-training minimizes:
 $$
 \mathcal{L}_{\mathrm{DPO}} = -\log \sigma\!\left(
@@ -124,7 +124,7 @@ The loss is adapted to the agent-environment boundary:
   next retrieval result.
 Both the initial policy and frozen reference use the pinned AetherSearch SFT
-checkpoint. The released run uses $\beta=0.1$.
 ## 🛠️ Training recipe

 Supervised fine-tuning teaches the model how to follow the search protocol;
 DPO then teaches it which of two valid-looking continuations is preferable.
+For a shared prompt \(x\), preferred continuation \(y_w\), rejected
+continuation \(y_l\), policy \(\pi_\theta\), and frozen SFT reference
+\(\pi_{\mathrm{ref}}\), training minimizes:
 $$
 \mathcal{L}_{\mathrm{DPO}} = -\log \sigma\!\left(
   next retrieval result.
 Both the initial policy and frozen reference use the pinned AetherSearch SFT
+checkpoint. The released run uses \(\beta=0.1\).
 ## 🛠️ Training recipe