Instructions to use Alexps9yyy/Bengal with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Alexps9yyy/Bengal with Transformers:

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

pipe = pipeline("text-generation", model="Alexps9yyy/Bengal")
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("Alexps9yyy/Bengal")
model = AutoModelForCausalLM.from_pretrained("Alexps9yyy/Bengal", 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 Alexps9yyy/Bengal with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Alexps9yyy/Bengal"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Alexps9yyy/Bengal",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Alexps9yyy/Bengal

SGLang

How to use Alexps9yyy/Bengal 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 "Alexps9yyy/Bengal" \
    --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": "Alexps9yyy/Bengal",
		"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 "Alexps9yyy/Bengal" \
        --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": "Alexps9yyy/Bengal",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Alexps9yyy/Bengal with Docker Model Runner:
```
docker model run hf.co/Alexps9yyy/Bengal
```

RLCR-QA (48-step probe)

Qwen3-8B belief scorer. Initialized from an Acc+Brier run (RLCR-D), then trained 48 GRPO steps on evidence-wrapped QA statements.

The model writes <think>, <analysis>, and <belief>. The belief is P(the wrapped statement is true).

Reward: 0.5 format + 0.5 Brier + 0.5 derived accuracy. This export is the 48-step checkpoint. A further 48 steps did not improve the QA evaluation.

On 149 held-out QA items with frozen retrieved evidence: AUROC 74.74, accuracy 73.15, Brier 18.62, ECE 8.26.

Downloads last month: 167

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for Alexps9yyy/Bengal

Base model

Qwen/Qwen3-8B-Base

Finetuned

Qwen/Qwen3-8B

Finetuned

(2,330)

this model